{ "@context": "https://schema.org", "@type": "Dataset", "dataset": { "name": "The AI for Science Landscape", "description": "Open dataset tracking companies, labs, finance intelligence, research evidence, milestones, and science-relevant AGI benchmarks driving the autonomous-science category.", "url": "https://scivity.org/landscape", "identifier": "https://scivity.org/landscape", "license": "https://creativecommons.org/licenses/by/4.0/", "isAccessibleForFree": true, "creator": { "@type": "Organization", "name": "Scivity Labs", "url": "https://scivity.org" }, "publisher": { "@type": "Organization", "name": "Scivity Labs", "url": "https://scivity.org" }, "citation": "Scivity Labs — AI for Science Landscape (https://scivity.org/landscape)", "spatialCoverage": "Global", "distribution": [ { "@type": "DataDownload", "encodingFormat": "application/json", "contentUrl": "https://scivity.org/landscape.json", "name": "Full landscape (JSON)" }, { "@type": "DataDownload", "encodingFormat": "text/csv", "contentUrl": "https://scivity.org/landscape.csv", "name": "Entity table (CSV)" }, { "@type": "DataDownload", "encodingFormat": "text/csv", "contentUrl": "https://scivity.org/landscape/funding-rounds.csv", "name": "Funding rounds (CSV)" }, { "@type": "DataDownload", "encodingFormat": "text/csv", "contentUrl": "https://scivity.org/landscape/public-financials.csv", "name": "Public financials (CSV)" }, { "@type": "DataDownload", "encodingFormat": "text/csv", "contentUrl": "https://scivity.org/landscape/research-outputs.csv", "name": "Research outputs (CSV)" }, { "@type": "DataDownload", "encodingFormat": "text/csv", "contentUrl": "https://scivity.org/landscape/benchmark-results.csv", "name": "Benchmark results (CSV)" }, { "@type": "DataDownload", "encodingFormat": "text/csv", "contentUrl": "https://scivity.org/landscape/deals-partnerships.csv", "name": "Deals and partnerships (CSV)" }, { "@type": "DataDownload", "encodingFormat": "text/csv", "contentUrl": "https://scivity.org/landscape/clinical-trials.csv", "name": "Clinical trials (CSV)" } ], "includedInDataCatalog": { "@type": "DataCatalog", "name": "Scivity Labs Open Research Datasets", "url": "https://scivity.org/landscape" }, "dateModified": "2026-06-01", "variableMeasured": [ "Entity autonomy score", "Execution score", "Verification score", "Scientific evidence score", "Capital", "Traction", "Public-company revenue", "Public-company R&D expense", "Public-company marketable securities", "Cash runway months", "Private funding round amount", "Non-dilutive award ceiling", "Research output provenance", "Measured benchmark result", "Clinical trial phase and status", "Evidence correction status" ], "keywords": [ "AI for science", "autonomous laboratory", "self-driving lab", "AI scientist", "autonomous discovery", "AGI for science" ] }, "license": "CC BY 4.0", "attribution": "Scivity Labs — AI for Science Landscape (https://scivity.org/landscape)", "entities": [ { "id": "periodic-labs", "name": "Periodic Labs", "tier": "tier1", "category": "Full-Stack Lab", "website": "https://periodic.com", "description": "Autonomous physical labs for materials and chemistry. AI scientists generate hypotheses; robotic labs run synthesis and characterization; results feed back as training data unavailable anywhere else.", "funding": "$300M seed", "valuation": "$1.5B", "hq": "San Francisco, CA", "founded": "2025", "investors": "Andreessen Horowitz, Nvidia, Bezos, Eric Schmidt, DST, Accel, Jeff Dean, Elad Gil", "differentiator": "Ex-OpenAI (Liam Fedus, co-creator of ChatGPT) and ex-DeepMind (Ekin Dogus Cubuk, GNoME). Building superconductors as first target.", "entityType": "Company", "stage": "Private scaleup", "region": "North America", "scienceDomains": [ "Materials", "Chemistry" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Autonomous materials labs", "AI scientist workflow", "Experimental data flywheel" ], "scores": { "autonomy": 4, "execution": 4, "verification": 3, "scientificEvidence": 3, "capital": 4, "traction": 2 }, "evidenceLevel": "Announced platform", "accessModel": "Proprietary", "tractionSignals": [ "$300M seed", "Frontier-lab founding team", "Materials-first autonomous lab buildout" ], "sourceIds": [ "official-periodic-labs" ] }, { "id": "lila-sciences", "name": "Lila Sciences", "tier": "tier1", "category": "Full-Stack Lab", "website": "https://lila.ai", "description": "AI Science Factories — integrated reasoning models, robotic labs, and verifiers running the scientific method across life, chemistry, and materials sciences.", "funding": "$550M", "valuation": "$1.3B", "hq": "Cambridge, MA", "founded": "2023", "investors": "Flagship Pioneering, Braidwell, Collective Global, General Catalyst, March, ARK, NVIDIA NVentures, ADIA", "differentiator": "Born out of Flagship Pioneering. George Church as chief scientist. Facilities in Boston, SF, London.", "entityType": "Company", "stage": "Private scaleup", "region": "North America", "scienceDomains": [ "Biology", "Chemistry", "Materials" ], "stackRoles": [ "Reason", "Execute", "Verify", "Data" ], "products": [ "AI Science Factories", "Reasoning models", "Robotic lab network" ], "scores": { "autonomy": 4, "execution": 4, "verification": 3, "scientificEvidence": 3, "capital": 4, "traction": 2 }, "evidenceLevel": "Announced platform", "accessModel": "Proprietary", "tractionSignals": [ "$550M raised", "Flagship Pioneering origin", "Life, chemical, and materials scope" ], "sourceIds": [ "official-lila-sciences" ] }, { "id": "sakana-ai", "name": "Sakana AI", "tier": "tier1", "category": "Autonomous Researcher", "website": "https://sakana.ai", "description": "The AI Scientist — autonomous ML research agent. Produces end-to-end research papers with minimal human involvement.", "funding": "$379M cumulative", "valuation": "$2.65B", "hq": "Tokyo, Japan", "founded": "2023", "investors": "NEA, Khosla, Lux Capital, NVIDIA, MUFG, In-Q-Tel", "differentiator": "Japan's most valuable private AI startup. Co-founded by David Ha and Llion Jones (Transformer paper).", "entityType": "Company", "stage": "Private scaleup", "region": "Asia-Pacific", "scienceDomains": [ "ML research", "General science" ], "stackRoles": [ "Reason", "Verify" ], "products": [ "AI Scientist", "AI Scientist-v2", "Marlin" ], "scores": { "autonomy": 4, "execution": 1, "verification": 2, "scientificEvidence": 3, "capital": 4, "traction": 3 }, "evidenceLevel": "Peer-reviewed demo", "accessModel": "Hybrid", "tractionSignals": [ "AI-authored paper workflow", "$379M cumulative funding", "Commercial enterprise research beta" ], "sourceIds": [ "official-sakana-ai" ] }, { "id": "futurehouse", "name": "FutureHouse", "tier": "tier1", "category": "Autonomous Researcher", "website": "https://futurehouse.org", "description": "Nonprofit AI research lab building autonomous scientists for biology. Released the Crow, Falcon, Owl, and Phoenix agent family, added ether0 and DISCO in 2026, and spun out Edison Scientific in late 2025.", "funding": "~$20M grants + $70M spinout seed", "hq": "San Francisco, CA", "founded": "2023", "investors": "Eric Schmidt, Schmidt Futures; Edison spinout led by Spark Capital", "differentiator": "Biology-first multi-agent system spanning literature, reasoning, and wet-lab-linked model work. DISCO and other 2026 releases show the nonprofit still moving the research frontier while Edison commercializes the stack.", "entityType": "Nonprofit lab", "stage": "Research platform", "region": "North America", "scienceDomains": [ "Biology", "Literature", "General science" ], "stackRoles": [ "Reason", "Execute", "Verify", "Data" ], "products": [ "Crow", "Falcon", "Owl", "Phoenix", "PaperQA2", "Robin", "ether0", "DISCO" ], "scores": { "autonomy": 4, "execution": 1, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Open research", "accessModel": "Hybrid", "tractionSignals": [ "Open scientific agents", "LitQA2 performance", "DISCO wet-lab validation", "Commercial spinout via Edison Scientific" ], "sourceIds": [ "official-futurehouse", "official-futurehouse-disco-202604" ] }, { "id": "autoscience-institute", "name": "Autoscience Institute", "tier": "tier1", "category": "Autonomous Researcher", "website": "https://autoscience.ai", "description": "Autonomous agents for ML research. Carl ideates and writes papers; Mira embeds new research directly into production ML models.", "funding": "$14M seed", "hq": "San Francisco, CA", "founded": "2023", "investors": "General Catalyst, Toyota Ventures, Perplexity Fund, S32", "differentiator": "Carl was the first AI system to generate a full-length peer-reviewed workshop paper (ICLR 2025, 3 of 4 submissions accepted).", "entityType": "Company", "stage": "Seed", "region": "North America", "scienceDomains": [ "ML research" ], "stackRoles": [ "Reason", "Verify" ], "products": [ "Carl", "Mira" ], "scores": { "autonomy": 4, "execution": 1, "verification": 2, "scientificEvidence": 3, "capital": 2, "traction": 2 }, "evidenceLevel": "Peer-reviewed demo", "accessModel": "Proprietary", "tractionSignals": [ "$14M seed", "ICLR workshop acceptances", "Live Kaggle competition entries" ], "sourceIds": [ "official-autoscience-institute" ] }, { "id": "thesis", "name": "Thesis", "tier": "tier1", "category": "Autonomous Researcher", "website": "https://www.ycombinator.com/companies/thesis", "description": "Applied research lab automating parts of the ML research workflow — experiment planning, training, evaluation, iteration.", "funding": "~$3M", "hq": "San Francisco, CA", "founded": "2025", "investors": "Y Combinator (F25 batch)", "differentiator": "State-of-the-art on OpenAI MLE-Bench. Founded by Stanford AI Lab alumni working with Chelsea Finn and Andrew Ng.", "entityType": "Company", "stage": "Seed", "region": "North America", "scienceDomains": [ "ML research" ], "stackRoles": [ "Reason", "Verify" ], "products": [ "Automated ML research workflow" ], "scores": { "autonomy": 3, "execution": 1, "verification": 2, "scientificEvidence": 2, "capital": 1, "traction": 1 }, "evidenceLevel": "Benchmark result", "accessModel": "Proprietary", "tractionSignals": [ "YC F25", "MLE-Bench medal-rate claim", "Stanford AI Lab founding network" ], "sourceIds": [ "official-thesis" ] }, { "id": "zeon-systems", "name": "Zeon Systems", "tier": "tier1", "category": "Lab Robotics", "website": "https://zeonsystems.ai", "description": "Natural-language-to-robot-arm layer for lab automation. Scientists type protocols in English; software programs off-the-shelf robotics to execute them.", "funding": "$0.5M", "hq": "San Francisco, CA", "founded": "2025", "investors": "Y Combinator (X25), FCVC, A* Capital", "differentiator": "Pilots at UCSF and Stanford. Computer-vision labware detection. Positions itself as the 'action endpoint' for AI science agents.", "entityType": "Company", "stage": "Seed", "region": "North America", "scienceDomains": [ "Lab automation", "Biology" ], "stackRoles": [ "Execute", "Infrastructure" ], "products": [ "Natural-language lab robotics layer", "Labware vision system" ], "scores": { "autonomy": 2, "execution": 3, "verification": 1, "scientificEvidence": 1, "capital": 1, "traction": 1 }, "evidenceLevel": "Pilot", "accessModel": "Proprietary", "tractionSignals": [ "YC X25", "UCSF and Stanford pilots", "Robot-arm protocol execution" ], "sourceIds": [ "official-zeon-systems" ] }, { "id": "automata", "name": "Automata", "tier": "tier2", "category": "Lab Robotics", "website": "https://automata.tech", "description": "Lab-automation operating system for AI-ready wet labs. Modular robotics, orchestration software, and unified data infrastructure to turn physical labs into programmable, repeatable systems.", "funding": "$45M Series C", "hq": "London, UK", "founded": "2015", "investors": "Dimension (lead), Danaher Ventures, Tru Arrow Partners, Octopus Ventures, Entrepreneurs First", "differentiator": "Pharma-first go-to-market — five of the top pharma companies are customers. Danaher strategic partnership on the Series C, with a board seat.", "entityType": "Company", "stage": "Growth", "region": "Europe", "scienceDomains": [ "Biology", "Diagnostics", "Lab automation" ], "stackRoles": [ "Execute", "Infrastructure", "Data" ], "products": [ "LINQ", "Automated lab benches", "LINQ Cloud" ], "scores": { "autonomy": 2, "execution": 4, "verification": 2, "scientificEvidence": 2, "capital": 2, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "Five top pharma customers", "$45M Series C", "Danaher strategic participation" ], "sourceIds": [ "official-automata" ] }, { "id": "cenevo", "name": "Cenevo", "tier": "tier2", "category": "Lab Orchestration / Lab Informatics", "website": "https://www.cenevo.com", "description": "Lab orchestration and informatics platform formed from Labguru and Titian Software. Provides ELN/LIMS, sample management, workflow automation, protocol conversion, and AI agents for regulated labs.", "funding": "Private", "hq": "United States / United Kingdom", "founded": "2025 rebrand", "investors": "Battery Ventures", "differentiator": "Commercial informatics footprint with 950+ customer organizations, 45,000+ users, and 8 of the top 10 pharma companies, plus agentic orchestration features with approvals, audit trails, and GxP positioning.", "entityType": "Company", "stage": "Commercial", "region": "North America / Europe", "scienceDomains": [ "Biology", "Lab automation", "Scientific informatics" ], "stackRoles": [ "Execute", "Data", "Infrastructure" ], "products": [ "Labguru ELN/LIMS", "Titian Mosaic sample management", "AI Protocol Conversion", "AI Automation agent" ], "scores": { "autonomy": 2, "execution": 3, "verification": 2, "scientificEvidence": 2, "capital": 2, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "950+ customer organizations", "45,000+ users", "8 of the top 10 pharma companies" ], "sourceIds": [ "official-cenevo-rebrand-202507", "rdworld-cenevo-ai-agents-202602" ] }, { "id": "emerald-cloud-lab", "name": "Emerald Cloud Lab", "tier": "tier2", "category": "Cloud Lab", "website": "https://www.emeraldcloudlab.com", "description": "Remote-controlled cloud lab — programmable chemistry and biology infrastructure accessed via a full scientific-computing environment.", "funding": "Private, undisclosed", "hq": "South San Francisco, CA", "founded": "2010", "investors": "Private", "differentiator": "Most mature programmable cloud lab. Used by pharma, Emerald Therapeutics, and academic groups at CMU.", "entityType": "Company", "stage": "Commercial", "region": "North America", "scienceDomains": [ "Chemistry", "Biology", "Cloud lab" ], "stackRoles": [ "Execute", "Infrastructure", "Data", "Verify" ], "products": [ "Remote programmable lab", "Scientific computing environment" ], "scores": { "autonomy": 2, "execution": 4, "verification": 3, "scientificEvidence": 3, "capital": 2, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "Mature programmable cloud lab", "Academic and pharma users", "Remote experiment execution" ], "sourceIds": [ "official-emerald-cloud-lab" ] }, { "id": "strateos", "name": "Strateos", "tier": "tier2", "category": "Cloud Lab", "website": "https://strateos.com", "description": "Robotic cloud lab for life sciences. Automated workflows for drug discovery, synthesis, and assays, accessible via API.", "funding": "$56M", "hq": "San Diego, CA", "founded": "2012", "investors": "Google Ventures, Mubadala", "differentiator": "Pharmaceutical R&D focus. API-first design for remote experiment orchestration.", "entityType": "Company", "stage": "Commercial", "region": "North America", "scienceDomains": [ "Biology", "Chemistry", "Cloud lab" ], "stackRoles": [ "Execute", "Infrastructure", "Data" ], "products": [ "Robotic cloud lab", "API experiment orchestration", "On-prem deployments" ], "scores": { "autonomy": 2, "execution": 4, "verification": 2, "scientificEvidence": 2, "capital": 2, "traction": 3 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "Pharma R&D focus", "$56M funding", "API-first orchestration" ], "sourceIds": [ "official-strateos" ] }, { "id": "chemify", "name": "Chemify", "tier": "tier2", "category": "Programmable Chemistry", "website": "https://chemify.io", "description": "Chemputer platform — universal chemical synthesis via programmable hardware and AI-driven reaction planning. Commercialization of Lee Cronin's Glasgow work.", "funding": "£43M", "hq": "Glasgow, UK", "founded": "2023", "investors": "Triatomic Capital, DCVC", "differentiator": "Universal programmable synthesis. Chemistry-as-code academic pedigree (Cronin lab).", "entityType": "Company", "stage": "Growth", "region": "Europe", "scienceDomains": [ "Chemistry", "Drug discovery", "Programmable synthesis" ], "stackRoles": [ "Reason", "Execute", "Verify", "Infrastructure" ], "products": [ "Chemputer", "Chemifarm", "Reaction planning software" ], "scores": { "autonomy": 3, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 3, "traction": 3 }, "evidenceLevel": "Published platform", "accessModel": "Proprietary", "tractionSignals": [ "Chemistry-as-code pedigree", "Series B expansion", "Peer-reviewed validation" ], "sourceIds": [ "official-chemify" ] }, { "id": "recursion-pharmaceuticals", "name": "Recursion Pharmaceuticals", "tier": "tier3", "category": "AI-Native Biotech", "website": "https://www.recursion.com", "description": "Publicly traded AI-native drug discovery platform. Industrial-scale phenomics, ML-driven target identification, and image-based screening at PB-scale.", "funding": "Public (NASDAQ: RXRX)", "valuation": "~$2.5B", "hq": "Salt Lake City, UT", "founded": "2013", "investors": "Public market", "differentiator": "Largest industrial AI biology platform. Partnerships with Bayer, Roche, Nvidia. Merged with Exscientia in 2024.", "entityType": "Public company", "stage": "Public", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery", "Imaging" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Recursion OS", "Industrial phenomics platform", "BioHive supercomputer" ], "scores": { "autonomy": 2, "execution": 3, "verification": 3, "scientificEvidence": 4, "capital": 4, "traction": 4 }, "evidenceLevel": "Public company", "accessModel": "Proprietary", "tractionSignals": [ "NASDAQ-listed", "Exscientia merger", "Industrial-scale image-based screening" ], "sourceIds": [ "official-recursion-pharmaceuticals" ] }, { "id": "atinary", "name": "Atinary", "tier": "tier3", "category": "Self-Driving Lab Software", "website": "https://atinary.com", "description": "Self-driving lab orchestration and execution layer. Runs SDLabs software plus physical Scientific Discovery Factories for chemistry, materials, and pharma R&D teams.", "funding": "Undisclosed", "hq": "Lausanne, Switzerland", "founded": "2019", "investors": "Private", "differentiator": "Commercial hardware-and-software self-driving lab stack that connects no-code AI optimization with robotics and analytical instruments.", "entityType": "Company", "stage": "Commercial", "region": "Europe", "scienceDomains": [ "Chemistry", "Materials", "Drug discovery" ], "stackRoles": [ "Reason", "Execute", "Verify", "Data" ], "products": [ "Self-driving lab software", "Boston Scientific Discovery Factories", "Bayesian optimization workflows" ], "scores": { "autonomy": 3, "execution": 3, "verification": 3, "scientificEvidence": 3, "capital": 1, "traction": 2 }, "evidenceLevel": "Commercial hardware and software", "accessModel": "Proprietary", "tractionSignals": [ "Software-first self-driving lab approach", "Boston Scientific Discovery Factories announced in 2026", "Chemistry informatics partnerships", "Enterprise R&D focus" ], "sourceIds": [ "official-atinary", "official-atinary-boston-sdl-202602", "selectscience-atinary-boston-sdl-202602" ] }, { "id": "ginkgo-bioworks", "name": "Ginkgo Bioworks", "tier": "tier3", "category": "Platform Biotech", "website": "https://www.ginkgobioworks.com", "description": "Synthetic biology foundry. Engineered organisms as a service; increasingly AI-augmented across the design-build-test-learn cycle.", "funding": "Public (NYSE: DNA)", "valuation": "~$1B", "hq": "Boston, MA", "founded": "2008", "investors": "Public market", "differentiator": "Largest organism engineering platform. Scale advantage in biological data generation.", "entityType": "Public company", "stage": "Public", "region": "North America", "scienceDomains": [ "Biology", "Synthetic biology", "Cloud lab" ], "stackRoles": [ "Reason", "Execute", "Data", "Infrastructure" ], "products": [ "Foundry", "Nebula", "Ginkgo Cloud Lab", "EstiMate" ], "scores": { "autonomy": 3, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 3, "traction": 4 }, "evidenceLevel": "Public company", "accessModel": "Hybrid", "tractionSignals": [ "NYSE-listed", "Nebula autonomous lab", "Large biological data generation platform" ], "sourceIds": [ "official-ginkgo-bioworks" ] }, { "id": "radical-ai", "name": "Radical AI", "tier": "tier3", "category": "AI-Native Materials", "website": "https://radicalai.com", "description": "AI platform for materials discovery. Focus on industrial biomanufacturing and novel functional materials.", "funding": "$13.5M", "hq": "New York, NY", "founded": "2022", "investors": "Plural, Y Combinator", "differentiator": "Materials-focused foundation models. Commercialization via industrial partnerships.", "entityType": "Company", "stage": "Seed", "region": "North America", "scienceDomains": [ "Materials", "Simulation" ], "stackRoles": [ "Reason", "Execute", "Data" ], "products": [ "Materials foundation models", "TorchSim", "Autonomous materials lab" ], "scores": { "autonomy": 3, "execution": 3, "verification": 2, "scientificEvidence": 2, "capital": 2, "traction": 2 }, "evidenceLevel": "Open research", "accessModel": "Hybrid", "tractionSignals": [ "TorchSim open-source release", "AFWERX contract", "Brooklyn autonomous materials lab" ], "sourceIds": [ "official-radical-ai" ] }, { "id": "xaira-therapeutics", "name": "Xaira Therapeutics", "tier": "tier3", "category": "AI-Native Biotech", "website": "https://www.xaira.com", "description": "AI-driven drug discovery and design company. Uses ML from hypothesis through clinical development.", "funding": "$1B+ Series A", "hq": "South San Francisco, CA", "founded": "2024", "investors": "ARCH Venture Partners, Foresite Labs, F-Prime, NEA, Sequoia, Lux, Lightspeed", "differentiator": "Largest initial biotech commitment in ARCH history. Co-founded by Nobel laureate David Baker and Marc Tessier-Lavigne.", "entityType": "Company", "stage": "Private scaleup", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery", "Virtual cell" ], "stackRoles": [ "Reason", "Data", "Verify" ], "products": [ "X-Atlas", "X-Cell", "AI drug discovery pipeline" ], "scores": { "autonomy": 2, "execution": 2, "verification": 3, "scientificEvidence": 3, "capital": 4, "traction": 2 }, "evidenceLevel": "Published data asset", "accessModel": "Proprietary", "tractionSignals": [ "$1B+ Series A", "25.6M-cell atlas", "Nobel laureate founder network" ], "sourceIds": [ "official-xaira-therapeutics" ] }, { "id": "helical", "name": "Helical", "tier": "tier3", "category": "AI-Native Biotech", "website": "https://www.helical.bio", "description": "Virtual AI lab for pharma R&D. Converts biological foundation models into reproducible in-silico discovery workflows for target identification, biomarker discovery, and therapeutic design.", "funding": "$10M seed", "valuation": "undisclosed", "hq": "London, UK", "founded": "2024", "investors": "redalpine, Gradient, BoxGroup, Frst, Aidan Gomez, Clement Delangue", "differentiator": "Production deployments with top-20 pharma teams, including a public Pfizer biomarker collaboration; bridges biologists and ML engineers through shared virtual-lab and model-factory surfaces.", "entityType": "Company", "stage": "Seed", "region": "Europe", "scienceDomains": [ "Biology", "Drug discovery", "Biomarkers" ], "stackRoles": [ "Reason", "Data", "Verify" ], "products": [ "Virtual Lab", "Model Factory", "Bio foundation model workflows" ], "scores": { "autonomy": 3, "execution": 1, "verification": 3, "scientificEvidence": 2, "capital": 2, "traction": 3 }, "evidenceLevel": "Production deployment", "accessModel": "Proprietary", "tractionSignals": [ "$10M seed", "Production deployments with multiple top-20 pharma companies", "Public Pfizer collaboration on predictive blood-based safety biomarkers" ], "sourceIds": [ "official-helical", "globenewswire-helical-seed-202604", "techeu-helical-seed-202604" ] }, { "id": "fathom-therapeutics", "name": "Fathom Therapeutics", "tier": "tier3", "category": "AI-Native Biotech", "website": "https://www.fathomtx.com", "description": "Physics- and AI-driven small-molecule drug design company formerly known as Atommap. Microcosmos simulates protein motion and interactions at atomic resolution to guide lab-in-the-loop discovery.", "funding": "$47M Series A", "valuation": "undisclosed", "hq": "New York, NY", "founded": "2023", "investors": "Sutter Hill Ventures, Chemistry, Alexandria Venture Investments, Empire State Development's NY Ventures", "differentiator": "Founded by Huafeng Xu, Yujie Wu, and Jesus Izaguirre; combines quantum chemistry, molecular dynamics, and AI to model dynamic protein behavior rather than static structures.", "entityType": "Company", "stage": "Series A", "region": "North America", "scienceDomains": [ "Drug discovery", "Chemistry", "Biology" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Microcosmos", "AI-enabled small-molecule discovery engine", "Lab-in-the-loop discovery programs" ], "scores": { "autonomy": 2, "execution": 2, "verification": 2, "scientificEvidence": 3, "capital": 3, "traction": 3 }, "evidenceLevel": "Working prototype", "accessModel": "Proprietary", "tractionSignals": [ "$47M Series A", "Lead program in animal efficacy studies", "Multiple external discovery partnerships" ], "sourceIds": [ "official-fathom-therapeutics", "prnewswire-fathom-series-a-202604", "cen-fathom-series-a-202604" ] }, { "id": "10x-science", "name": "10x Science", "tier": "tier4", "category": "AI-Native Biotech", "website": "https://10xscience.com", "description": "AI-native platform for next-generation protein characterization. Automates mass-spectrometry interpretation so biologic drug candidates can be characterized and quality-assessed faster.", "funding": "$4.8M seed", "valuation": "undisclosed", "hq": "San Francisco, CA", "founded": "2025", "investors": "Initialized Capital, Y Combinator, Civilization Ventures, Founder Factor", "differentiator": "Stanford/Bertozzi-lab founding team attacking the verification bottleneck created by AI-generated biologics candidates.", "entityType": "Company", "stage": "Seed", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery", "Protein characterization" ], "stackRoles": [ "Data", "Verify" ], "products": [ "AI-native protein characterization platform", "Mass spectrometry analysis agents", "Proteoform analysis workflows" ], "scores": { "autonomy": 2, "execution": 1, "verification": 3, "scientificEvidence": 2, "capital": 1, "traction": 2 }, "evidenceLevel": "Announced platform", "accessModel": "Proprietary", "tractionSignals": [ "$4.8M seed", "YC W26", "Stanford/Bertozzi-lab founder expertise" ], "sourceIds": [ "official-10x-science", "yc-10x-science-profile-202604", "techcrunch-10x-science-seed-202604", "prnewswire-10x-science-seed-202604" ] }, { "id": "a-lab-lbnl", "name": "A-Lab (LBNL)", "tier": "tier4", "category": "National Lab Facility", "website": "https://a-lab.lbl.gov", "description": "Autonomous materials synthesis lab at Lawrence Berkeley National Lab. AI-driven synthesis of novel inorganic materials.", "funding": "Government", "hq": "Berkeley, CA", "founded": "2022", "investors": "US Department of Energy", "differentiator": "Reference implementation for autonomous materials labs. Nature-published 2023; subject of a major 2024 critique and 2026 correction — the defining verification case study of the field.", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Materials", "Chemistry" ], "stackRoles": [ "Reason", "Execute", "Verify", "Data" ], "products": [ "Autonomous materials synthesis facility" ], "scores": { "autonomy": 4, "execution": 4, "verification": 2, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Published platform", "accessModel": "Public infrastructure", "tractionSignals": [ "Nature-published autonomous synthesis", "DOE national lab context", "Defining verification case study" ], "sourceIds": [ "official-a-lab-lbnl" ] }, { "id": "acceleration-consortium", "name": "Acceleration Consortium", "tier": "tier4", "category": "Academic Network", "website": "https://acceleration.utoronto.ca", "description": "International consortium based at University of Toronto. Coordinates self-driving labs across member institutions; publishes protocols and standards.", "funding": "$200M+ (CFREF grant)", "hq": "Toronto, Canada", "founded": "2021", "investors": "Canada First Research Excellence Fund", "differentiator": "Standards body for the autonomous-lab field. Over 30 member institutions worldwide.", "entityType": "Academic network", "stage": "Research network", "region": "North America", "scienceDomains": [ "Materials", "Chemistry", "Standards" ], "stackRoles": [ "Execute", "Verify", "Infrastructure", "Data" ], "products": [ "Self-driving lab network", "Protocols", "Standards coordination" ], "scores": { "autonomy": 2, "execution": 3, "verification": 3, "scientificEvidence": 4, "capital": 3, "traction": 3 }, "evidenceLevel": "Research network", "accessModel": "Public infrastructure", "tractionSignals": [ "$200M CFREF grant", "30+ member institutions", "Standards role" ], "sourceIds": [ "official-acceleration-consortium" ] }, { "id": "pnnl", "name": "PNNL", "tier": "tier4", "category": "National Lab", "website": "https://www.pnnl.gov", "description": "Pacific Northwest National Laboratory. DOE lab with autonomous experimentation programs in chemistry and energy. Part of the DOE × DeepMind Genesis Mission network.", "funding": "Government", "hq": "Richland, WA", "founded": "1965", "investors": "US Department of Energy", "differentiator": "National-lab infrastructure. Deep expertise in autonomous chemistry and energy systems.", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Chemistry", "Energy", "Materials" ], "stackRoles": [ "Execute", "Verify", "Infrastructure", "Data" ], "products": [ "Autonomy Studio", "DOE autonomous experimentation programs" ], "scores": { "autonomy": 2, "execution": 3, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Research facility", "accessModel": "Public infrastructure", "tractionSignals": [ "DOE national lab", "Autonomy Studio", "Genesis Mission network" ], "sourceIds": [ "official-pnnl" ] }, { "id": "google-deepmind", "name": "Google DeepMind", "tier": "tier5", "category": "Frontier AI Lab", "website": "https://deepmind.google", "description": "AlphaFold, AlphaProteo, GNoME, AlphaGeometry, AlphaEvolve — state-of-the-art scientific AI, without a dedicated autonomous-lab product line.", "funding": "Alphabet subsidiary", "hq": "London, UK", "founded": "2010", "investors": "Alphabet", "differentiator": "Deepest scientific AI bench in the industry. Genesis Mission partnership with DOE deploys Gemini across 17 US National Labs.", "entityType": "Frontier AI lab", "stage": "Big Tech lab", "region": "Europe", "scienceDomains": [ "Biology", "Materials", "Mathematics", "General science" ], "stackRoles": [ "Reason", "Verify", "Data" ], "products": [ "AlphaFold", "AlphaProteo", "GNoME", "AlphaGeometry", "AI co-scientist" ], "scores": { "autonomy": 3, "execution": 1, "verification": 3, "scientificEvidence": 4, "capital": 4, "traction": 4 }, "evidenceLevel": "Frontier research", "accessModel": "Hybrid", "tractionSignals": [ "Nobel-recognized AlphaFold work", "DOE Genesis Mission partnership", "Deep science AI bench" ], "sourceIds": [ "official-google-deepmind", "epoch-capabilities" ] }, { "id": "openai", "name": "OpenAI", "tier": "tier5", "category": "Frontier AI Lab", "website": "https://openai.com", "description": "Frontier model developer. April 2026 Industrial Policy paper explicitly calls for distributed AI-enabled laboratories as public infrastructure.", "funding": "$40B+ raised", "valuation": "$500B", "hq": "San Francisco, CA", "founded": "2015", "investors": "Microsoft, Thrive, Founders Fund", "differentiator": "Deep Research agent. Explicit policy focus on autonomous science, though no dedicated science product yet.", "entityType": "Frontier AI lab", "stage": "Frontier platform", "region": "North America", "scienceDomains": [ "General science", "Biology", "Reasoning" ], "stackRoles": [ "Reason", "Verify", "Infrastructure" ], "products": [ "Deep Research", "o-series reasoning models", "GPT-Rosalind", "Prism" ], "scores": { "autonomy": 3, "execution": 1, "verification": 3, "scientificEvidence": 3, "capital": 4, "traction": 4 }, "evidenceLevel": "Frontier platform", "accessModel": "Proprietary", "tractionSignals": [ "DOE national labs access", "Science policy agenda", "Specialized biology model launch" ], "sourceIds": [ "official-openai", "artificial-analysis-leaderboard", "artificial-analysis-methodology" ] }, { "id": "isomorphic-labs", "name": "Isomorphic Labs", "tier": "tier3", "category": "AI-Native Biotech", "website": "https://www.isomorphiclabs.com", "description": "Alphabet-backed AI drug discovery company spun out of Google DeepMind. Builds on AlphaFold to design novel therapeutics, with a 17-program pipeline advancing toward the clinic.", "funding": "$600M (first external round)", "valuation": "Undisclosed", "hq": "London, UK", "founded": "2021", "investors": "Thrive Capital, GV, Alphabet", "differentiator": "Founded and led by Demis Hassabis. Nearly $3B in collaborations with Eli Lilly and Novartis. IsoDDE design engine surpasses AlphaFold 3 at protein-ligand prediction; first AI-designed oncology drug targeting Phase 1 trials in 2026.", "entityType": "Company", "stage": "Private scaleup", "region": "Europe", "scienceDomains": [ "Biology", "Drug discovery" ], "stackRoles": [ "Reason", "Verify", "Data" ], "products": [ "IsoDDE", "AI drug design platform" ], "scores": { "autonomy": 2, "execution": 2, "verification": 3, "scientificEvidence": 4, "capital": 4, "traction": 3 }, "evidenceLevel": "Clinical pipeline", "accessModel": "Proprietary", "tractionSignals": [ "$600M first external round", "Nearly $3B in pharma collaborations", "17-program pipeline" ], "sourceIds": [ "official-isomorphic-labs" ] }, { "id": "edison-scientific", "name": "Edison Scientific", "tier": "tier1", "category": "Autonomous Researcher", "website": "https://edisonscientific.com", "description": "Commercial spinout of FutureHouse. Kosmos is a multi-agent AI co-scientist that runs 12-hour cycles of literature review, data analysis, and hypothesis generation — executing ~42,000 lines of code and reading ~1,500 papers per run. Its 2026 surface also includes PaperQA3-backed Edison Literature and LABBench2.", "funding": "$70M seed", "valuation": "$250M", "hq": "San Francisco, CA", "founded": "2025", "investors": "Spark Capital, Triatomic Capital; angels include Jeff Dean and Dmitri Alperovitch", "differentiator": "CEO Sam Rodriques and co-founder Andrew White. Kosmos launch paper (arXiv 2511.02824) demonstrated seven discoveries across neuroscience, materials, and statistical genetics — three reproducing human-led findings, four entirely novel — while 2026 updates deepened the platform's literature and benchmark layer.", "entityType": "Company", "stage": "Seed", "region": "North America", "scienceDomains": [ "Biology", "Materials", "Statistical genetics", "General science" ], "stackRoles": [ "Reason", "Verify", "Data" ], "products": [ "Kosmos", "Edison Literature", "PaperQA3", "LABBench2" ], "scores": { "autonomy": 4, "execution": 1, "verification": 3, "scientificEvidence": 3, "capital": 3, "traction": 2 }, "evidenceLevel": "Launch paper", "accessModel": "Proprietary", "tractionSignals": [ "$70M seed", "Seven demonstrated discoveries", "PaperQA3 literature upgrade", "NVIDIA scaling partnership", "FutureHouse commercial spinout" ], "sourceIds": [ "official-edison-scientific", "official-edison-paperqa3-202602", "official-edison-nvidia-202603", "official-edison-labbench2-202605" ] }, { "id": "insilico-medicine", "name": "Insilico Medicine", "tier": "tier3", "category": "AI-Native Biotech", "website": "https://insilico.com", "description": "End-to-end generative AI platform for drug discovery and development (Pharma.AI). Public on the Hong Kong Stock Exchange since December 2025. TNIK inhibitor rentosertib is the first AI-discovered and AI-designed drug to deliver Phase 2a clinical data.", "funding": "Public (HKG: 03696.HK)", "valuation": "~$2B", "hq": "Hong Kong (also Boston, New York)", "founded": "2014", "investors": "Lilly, Tencent, Temasek, Schroders, UBS, Oaktree, E Fund, Taikang Life", "differentiator": "First AI-driven biotech to list on HKEX. Largest Hong Kong biotech IPO of 2025, oversubscribed 1,427× retail. $2.75B Eli Lilly partnership plus $888M Servier oncology deal. Founded by Alex Zhavoronkov.", "entityType": "Public company", "stage": "Public", "region": "Asia-Pacific", "scienceDomains": [ "Biology", "Drug discovery", "Clinical development" ], "stackRoles": [ "Reason", "Verify", "Data" ], "products": [ "Pharma.AI", "rentosertib", "AI drug discovery pipeline" ], "scores": { "autonomy": 2, "execution": 2, "verification": 3, "scientificEvidence": 4, "capital": 4, "traction": 4 }, "evidenceLevel": "Clinical data", "accessModel": "Proprietary", "tractionSignals": [ "HKEX listing", "Phase 2a clinical data", "$2.75B Lilly partnership" ], "sourceIds": [ "official-insilico-medicine" ] }, { "id": "ami-labs", "name": "AMI Labs", "tier": "tier5", "category": "Frontier AI Lab", "website": "https://amilabs.xyz", "description": "Advanced Machine Intelligence — Yann LeCun's new lab. Building world models based on Joint Embedding Predictive Architecture (JEPA) rather than next-token LLMs; aims to learn from physical reality as the path to human-level intelligence.", "funding": "$1.03B seed", "valuation": "$3.5B pre-money", "hq": "Paris, France", "founded": "2025", "investors": "Bezos Expeditions, Cathay Innovation, Greycroft, Hiro Capital, HV Capital, Nvidia, Temasek, Daphni, SBVA", "differentiator": "Largest seed round in European tech history. Turing Award winner Yann LeCun at the helm after leaving Meta; Saining Xie as CSO, Pascale Fung as chief research officer, Michael Rabbat as VP of world models. Explicit bet against the LLM paradigm.", "entityType": "Frontier AI lab", "stage": "Seed", "region": "Europe", "scienceDomains": [ "World models", "General intelligence", "Physical reasoning" ], "stackRoles": [ "Reason", "Data" ], "products": [ "JEPA-first world model research agenda" ], "scores": { "autonomy": 2, "execution": 1, "verification": 1, "scientificEvidence": 1, "capital": 4, "traction": 1 }, "evidenceLevel": "Announced lab", "accessModel": "Proprietary", "tractionSignals": [ "$1.03B seed", "Yann LeCun founding team", "World-model-first thesis" ], "sourceIds": [ "official-ami-labs" ] }, { "id": "chemlex", "name": "ChemLex", "tier": "tier2", "category": "Programmable Chemistry", "website": "https://chemlex.com", "description": "AI-for-science startup running a 24/7 autonomous chemistry synthesis line. Serves as the chemistry discovery engine for pharma; global HQ and flagship self-driving lab in Singapore.", "funding": "~$71M (Series A + Series B)", "valuation": "Undisclosed", "hq": "Singapore", "founded": "2022", "investors": "Granite Asia (lead), Qiming Venture Partners, LYFE Capital, Sinovation Ventures", "differentiator": "70+ customers including six of the top ten global pharma companies. MoU with Singapore's Experimental Drug Development Centre (EDDC). Incubated from MegaRobo; first major AI-for-science lab anchored in Asia.", "entityType": "Company", "stage": "Growth", "region": "Asia-Pacific", "scienceDomains": [ "Chemistry", "Drug discovery", "Programmable synthesis" ], "stackRoles": [ "Reason", "Execute", "Infrastructure", "Data" ], "products": [ "24/7 autonomous synthesis line", "Chemistry CRO workflow" ], "scores": { "autonomy": 3, "execution": 4, "verification": 2, "scientificEvidence": 2, "capital": 3, "traction": 3 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "70+ customers", "Six top-ten pharma customers", "Singapore flagship self-driving lab" ], "sourceIds": [ "official-chemlex" ] }, { "id": "coscientist-cmu", "name": "Coscientist (CMU)", "tier": "tier2", "category": "Autonomous Researcher", "website": "https://www.nature.com/articles/s41586-023-06792-0", "description": "Academic autonomous chemistry system from Carnegie Mellon. Uses GPT-4 with tool use, code execution, documentation search, and cloud-lab automation to design, plan, and execute chemistry experiments.", "funding": "Academic research", "hq": "Pittsburgh, PA", "founded": "2023", "investors": "Carnegie Mellon University; Emerald Cloud Lab collaboration", "differentiator": "Landmark Nature demonstration of an LLM-driven chemistry agent running real palladium cross-coupling experiments through Emerald Cloud Lab.", "entityType": "Academic project", "stage": "Published research", "region": "North America", "scienceDomains": [ "Chemistry", "Autonomous agents", "Cloud lab" ], "stackRoles": [ "Reason", "Execute", "Verify", "Data" ], "products": [ "Coscientist autonomous chemistry system", "Cloud-lab experiment execution" ], "scores": { "autonomy": 4, "execution": 3, "verification": 3, "scientificEvidence": 4, "capital": 1, "traction": 2 }, "evidenceLevel": "Nature paper", "accessModel": "Research prototype", "tractionSignals": [ "Nature 2023 publication", "Emerald Cloud Lab execution", "Palladium cross-coupling demonstrations" ], "sourceIds": [ "cmu-coscientist-202312", "nature-coscientist-202312" ] }, { "id": "kebotix", "name": "Kebotix", "tier": "tier3", "category": "AI-Native Materials", "website": "https://www.kebotix.com", "description": "Self-driving laboratory company for chemicals and materials discovery. Combines AI planning, robotics, and high-throughput experimentation to shorten materials and molecule optimization loops.", "funding": "$11.4M Series A", "hq": "Cambridge, MA", "founded": "2017", "investors": "Novo Holdings, One Way Ventures, Baidu Ventures, Embark Ventures, Flybridge Capital", "differentiator": "Early commercial self-driving-lab pioneer focused on green chemistry and materials discovery rather than only informatics.", "entityType": "Company", "stage": "Commercial", "region": "North America", "scienceDomains": [ "Materials", "Chemistry", "Green chemistry" ], "stackRoles": [ "Reason", "Execute", "Data" ], "products": [ "Self-driving materials discovery platform", "AI-guided chemistry workflows" ], "scores": { "autonomy": 3, "execution": 3, "verification": 2, "scientificEvidence": 2, "capital": 2, "traction": 2 }, "evidenceLevel": "Commercial platform", "accessModel": "Proprietary", "tractionSignals": [ "$11.4M Series A", "Novo Holdings lead", "Self-driving-lab positioning" ], "sourceIds": [ "official-kebotix", "finsmes-kebotix-series-a-202004" ] }, { "id": "citrine-informatics", "name": "Citrine Informatics", "tier": "tier3", "category": "AI-Native Materials", "website": "https://citrine.io", "description": "Materials AI platform for chemicals, formulations, and manufactured products. Provides domain-specific data infrastructure and AI models that help R&D teams predict, optimize, and reuse materials knowledge.", "funding": "$43M+ disclosed", "hq": "Redwood City, CA", "founded": "2013", "investors": "Prelude Ventures, Innovation Endeavors, DCVC, Next47, Drive Catalyst", "differentiator": "One of the earliest enterprise materials-informatics platforms, with deployments at global chemicals and manufacturing companies.", "entityType": "Company", "stage": "Growth", "region": "North America", "scienceDomains": [ "Materials", "Chemistry", "Manufacturing" ], "stackRoles": [ "Reason", "Data", "Verify" ], "products": [ "Citrine Platform", "Materials AI models", "R&D data infrastructure" ], "scores": { "autonomy": 2, "execution": 1, "verification": 3, "scientificEvidence": 3, "capital": 3, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "$16M Series C", "Global chemicals customers", "Materials AI operating-platform positioning" ], "sourceIds": [ "official-citrine-informatics", "businesswire-citrine-series-c-202301" ] }, { "id": "artificial", "name": "Artificial", "tier": "tier3", "category": "Self-Driving Lab Software", "website": "https://www.artificial.com", "description": "Lab orchestration software for life sciences R&D. Connects wet-lab instruments, informatics systems, scheduling, digital twins, and AI agents into controlled scientific workflows.", "funding": "$21.5M Series A", "hq": "Palo Alto, CA", "founded": "2019", "investors": "Microsoft M12, Playground Global, AME Cloud Ventures", "differentiator": "Software-first control layer for robotic and human-operated labs, built to coordinate the full design-make-test-analyze cycle.", "entityType": "Company", "stage": "Series A", "region": "North America", "scienceDomains": [ "Lab automation", "Biology", "Synthetic biology" ], "stackRoles": [ "Execute", "Infrastructure", "Data" ], "products": [ "aLab Suite", "Lab orchestration platform", "Digital twin builder" ], "scores": { "autonomy": 3, "execution": 3, "verification": 2, "scientificEvidence": 2, "capital": 2, "traction": 2 }, "evidenceLevel": "Commercial platform", "accessModel": "Proprietary", "tractionSignals": [ "$21.5M Series A", "Thermo Fisher and Beam Therapeutics customers", "AI agents for lab process optimization" ], "sourceIds": [ "official-artificial", "techcrunch-artificial-series-a-202105" ] }, { "id": "synthace", "name": "Synthace", "tier": "tier3", "category": "Self-Driving Lab Software", "website": "https://www.synthace.com", "description": "Cloud experiment-design platform for biological R&D. Turns multivariate experiment designs into automation instructions and structured data that can be analyzed and reused.", "funding": "$35M Series C", "hq": "London, UK", "founded": "2011", "investors": "Horizons Ventures, Sofinnova Partners, SOSV, Bioeconomy Capital", "differentiator": "No-code bridge between design-of-experiments planning, liquid-handling automation, and AI-ready experimental data.", "entityType": "Company", "stage": "Growth", "region": "Europe", "scienceDomains": [ "Biology", "Experiment design", "Lab automation" ], "stackRoles": [ "Reason", "Execute", "Data" ], "products": [ "Synthace platform", "Automated multivariate experiments", "Automation instruction generation" ], "scores": { "autonomy": 2, "execution": 3, "verification": 2, "scientificEvidence": 2, "capital": 3, "traction": 3 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "$35M Series C", "Top pharma and CRO customers", "No-code DOE-to-automation workflow" ], "sourceIds": [ "official-synthace", "finsmes-synthace-series-c-202111" ] }, { "id": "tetrascience", "name": "TetraScience", "tier": "tier3", "category": "Self-Driving Lab Software", "website": "https://www.tetrascience.com", "description": "Scientific Data and AI Cloud for life sciences R&D. Replatforms raw instrument and experiment data into engineered, compliant datasets for analytics, automation, and scientific AI.", "funding": "$92M+ disclosed", "hq": "Boston, MA", "founded": "2014", "investors": "Insight Partners, Alkeon Capital, Waters Corporation", "differentiator": "Vendor-neutral data layer for AI-ready science, with broad lab-instrument and informatics integrations across enterprise pharma.", "entityType": "Company", "stage": "Growth", "region": "North America", "scienceDomains": [ "Scientific data", "Biopharma", "Lab automation" ], "stackRoles": [ "Data", "Infrastructure", "Verify" ], "products": [ "Tetra Scientific Data and AI Platform", "Tetra Data Platform", "Tetra AI" ], "scores": { "autonomy": 2, "execution": 1, "verification": 3, "scientificEvidence": 2, "capital": 3, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "$80M Series B", "Enterprise pharma adoption", "Instrument and informatics partner network" ], "sourceIds": [ "official-tetrascience", "fierce-tetrascience-series-b-202104" ] }, { "id": "riffyn", "name": "Riffyn", "tier": "tier4", "category": "Self-Driving Lab Software", "website": "https://plm.sw.siemens.com/en-US/riffyn-x/", "description": "Scientific process design and data platform now offered as Riffyn X inside Siemens. Captures experimental methods, sample lineage, and process data in forms ready for analysis and machine learning.", "funding": "$19M+ disclosed before Siemens/Dotmatics integration", "hq": "Oakland, CA", "founded": "2014", "investors": "M Ventures, Waters, Siemens Venture Capital, OATV", "differentiator": "Process-centric experiment model that links recipe design, execution, chain of custody, and machine-learning-ready data.", "entityType": "Company / product line", "stage": "Commercial", "region": "North America", "scienceDomains": [ "Bioprocessing", "Scientific data", "Experiment design" ], "stackRoles": [ "Data", "Verify", "Infrastructure" ], "products": [ "Riffyn X", "Scientific Development Environment", "Process data models" ], "scores": { "autonomy": 1, "execution": 1, "verification": 3, "scientificEvidence": 2, "capital": 2, "traction": 3 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "$15M Series B", "BASF collaboration", "Siemens / Dotmatics current product line" ], "sourceIds": [ "siemens-riffyn-x", "biospace-riffyn-series-b-201905" ] }, { "id": "opentrons", "name": "Opentrons", "tier": "tier3", "category": "Lab Robotics", "website": "https://opentrons.com", "description": "Open-source liquid-handling robotics company. Flex and OT-2 robots make programmable wet-lab automation accessible through modular hardware, Python APIs, and protocol libraries.", "funding": "Private, venture-backed", "hq": "New York, NY", "founded": "2013", "investors": "Khosla Ventures and others", "differentiator": "Low-cost, open robotics stack that lets smaller biology labs script and reproduce experiments without legacy automation overhead.", "entityType": "Company", "stage": "Commercial", "region": "North America", "scienceDomains": [ "Lab automation", "Biology", "Liquid handling" ], "stackRoles": [ "Execute", "Infrastructure" ], "products": [ "Opentrons Flex", "OT-2", "Python Protocol API", "Protocol Designer" ], "scores": { "autonomy": 2, "execution": 4, "verification": 2, "scientificEvidence": 3, "capital": 2, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Hybrid", "tractionSignals": [ "Open-source software and firmware", "Flex launch", "Mayo Clinic, Harvard, and MIT customer signals" ], "sourceIds": [ "official-opentrons-flex", "techcrunch-opentrons-flex-202305" ] }, { "id": "tecan", "name": "Tecan", "tier": "tier4", "category": "Lab Robotics", "website": "https://www.tecan.com", "description": "Swiss public lab-automation and life-science instrumentation company. Fluent and related workstations provide liquid handling, robotics, readers, and process automation across pharma and diagnostics.", "funding": "Public (SIX: TECN)", "hq": "Maennedorf, Switzerland", "founded": "1980", "investors": "Public market", "differentiator": "Industry-standard liquid-handling infrastructure that many AI-ready and self-driving lab stacks build around.", "entityType": "Public company", "stage": "Public", "region": "Europe", "scienceDomains": [ "Lab automation", "Diagnostics", "Life sciences" ], "stackRoles": [ "Execute", "Infrastructure" ], "products": [ "Fluent", "Freedom EVO", "Tecan Labwerx", "Liquid handling systems" ], "scores": { "autonomy": 1, "execution": 4, "verification": 2, "scientificEvidence": 3, "capital": 3, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Commercial hardware", "tractionSignals": [ "SIX-listed", "3,000+ employees", "Global lab automation installed base" ], "sourceIds": [ "official-tecan-history", "stockanalysis-tecan-profile-202603" ] }, { "id": "highres-biosolutions", "name": "HighRes Biosolutions", "tier": "tier3", "category": "Lab Robotics", "website": "https://www.highres.com", "description": "Life-science lab automation company combining modular hardware, liquid handling, dynamic scheduling, and Cellario orchestration software for drug discovery, genomics, and synthetic biology workflows.", "funding": "Private; Axel Johnson portfolio company", "hq": "Beverly, MA", "founded": "2004", "investors": "Axel Johnson", "differentiator": "Cellario and Nucleus turn heterogeneous instruments and workcells into schedulable, data-producing automation systems.", "entityType": "Company", "stage": "Commercial", "region": "North America", "scienceDomains": [ "Lab automation", "Drug discovery", "Synthetic biology", "Genomics" ], "stackRoles": [ "Execute", "Infrastructure", "Data" ], "products": [ "Cellario", "Nucleus", "Prime liquid handler", "AutoPod" ], "scores": { "autonomy": 2, "execution": 4, "verification": 2, "scientificEvidence": 2, "capital": 2, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Commercial hardware and software", "tractionSignals": [ "20+ years deployments", "Axel Johnson ownership", "Cellario OS orchestration platform" ], "sourceIds": [ "official-highres", "biodev-highres-cellario-202509" ] }, { "id": "generate-biomedicines", "name": "Generate Biomedicines", "tier": "tier2", "category": "AI-Native Biotech", "website": "https://generatebiomedicines.com", "description": "Generative biology company creating programmable protein therapeutics. Combines machine-learning design with biological engineering and high-throughput wet-lab validation to build a clinical pipeline.", "funding": "Nearly $700M private financing before IPO path", "hq": "Somerville, MA", "founded": "2018", "investors": "Flagship Pioneering, Amgen, NVIDIA NVentures, ARCH, T. Rowe Price, ADIA", "differentiator": "Flagship platform company using learned protein design to generate antibodies, peptides, enzymes, and other therapeutic proteins.", "entityType": "Company", "stage": "Clinical-stage platform", "region": "North America", "scienceDomains": [ "Biology", "Protein design", "Drug discovery" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Generate Platform", "Generative Biology", "Protein therapeutic pipeline" ], "scores": { "autonomy": 2, "execution": 3, "verification": 3, "scientificEvidence": 3, "capital": 4, "traction": 3 }, "evidenceLevel": "Clinical pipeline", "accessModel": "Proprietary", "tractionSignals": [ "$370M Series B", "$273M Series C", "Amgen collaboration", "17-program pipeline reported in 2023" ], "sourceIds": [ "official-generate-biomedicines", "fierce-generate-series-b-202111", "businesswire-generate-series-c-202309" ] }, { "id": "insitro", "name": "Insitro", "tier": "tier2", "category": "AI-Native Biotech", "website": "https://www.insitro.com", "description": "Machine-learning drug discovery company founded by Daphne Koller. Builds large biological datasets from in vitro disease models, human genetics, and partnerships to find targets and patient subgroups.", "funding": "$600M+ disclosed", "valuation": "~$2.5B reported", "hq": "South San Francisco, CA", "founded": "2018", "investors": "Andreessen Horowitz, ARCH, GV, BlackRock, Temasek, CPP Investments", "differentiator": "Combines automated biology, human genetics, and ML under one roof rather than selling a narrow model or CRO service.", "entityType": "Company", "stage": "Private scaleup", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery", "Human genetics" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Machine-learning drug discovery platform", "In vitro disease models", "Human genetics data platform" ], "scores": { "autonomy": 2, "execution": 3, "verification": 3, "scientificEvidence": 3, "capital": 4, "traction": 3 }, "evidenceLevel": "Commercial partnerships", "accessModel": "Proprietary", "tractionSignals": [ "Daphne Koller founder", "Lilly and Bristol Myers Squibb partnerships", "Reported $2.5B valuation" ], "sourceIds": [ "official-insitro-purpose", "ap-insitro-ai-drug-discovery-202412" ] }, { "id": "cellarity", "name": "Cellarity", "tier": "tier3", "category": "AI-Native Biotech", "website": "https://cellarity.com", "description": "Flagship Pioneering company developing cell-state-correcting medicines. Uses single-cell technologies, network biology, and machine learning to map disease cell behavior and design interventions.", "funding": "$274M disclosed", "hq": "Somerville, MA", "founded": "2017", "investors": "Flagship Pioneering, Kyowa Kirin, Hanwha Impact Partners", "differentiator": "Platform shifts discovery from single targets to whole-cell state transitions, with AI and multi-omics as the operating substrate.", "entityType": "Company", "stage": "Private scaleup", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery", "Single-cell biology" ], "stackRoles": [ "Reason", "Data", "Verify" ], "products": [ "Cell State-Correcting therapeutics platform", "Single-cell and ML discovery workflows" ], "scores": { "autonomy": 2, "execution": 2, "verification": 3, "scientificEvidence": 2, "capital": 3, "traction": 2 }, "evidenceLevel": "Platform pipeline", "accessModel": "Proprietary", "tractionSignals": [ "$121M Series C", "$274M total disclosed", "Flagship Pioneering origin" ], "sourceIds": [ "official-cellarity-about", "cellarity-series-c-202210" ] }, { "id": "abcellera", "name": "AbCellera", "tier": "tier3", "category": "Platform Biotech", "website": "https://www.abcellera.com", "description": "Public antibody-discovery platform company. Integrates single-cell screening, computation, antibody engineering, translational science, and clinical manufacturing from target to clinic.", "funding": "Public (NASDAQ: ABCL)", "hq": "Vancouver, Canada", "founded": "2012", "investors": "Public market; DARPA and pharma partnerships", "differentiator": "Full-stack antibody engine with in-house clinical manufacturing and a pandemic-response track record with Eli Lilly.", "entityType": "Public company", "stage": "Public", "region": "North America", "scienceDomains": [ "Biology", "Antibody discovery", "Clinical manufacturing" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Antibody discovery platform", "Single-cell screening", "Clinical manufacturing facility" ], "scores": { "autonomy": 2, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 4, "traction": 4 }, "evidenceLevel": "Public company", "accessModel": "Partnership platform", "tractionSignals": [ "NASDAQ-listed", "$555.5M IPO proceeds", "DARPA pandemic-response platform", "Eli Lilly COVID antibody collaboration" ], "sourceIds": [ "official-abcellera-platform", "renaissance-abcellera-ipo-202011" ] }, { "id": "iambic-therapeutics", "name": "Iambic Therapeutics", "tier": "tier3", "category": "AI-Native Biotech", "website": "https://www.iambic.ai", "description": "AI-driven small-molecule drug discovery company. Uses multimodal transformers, protein-ligand structure prediction, high-throughput chemistry, and weekly biological data loops to advance candidates.", "funding": "$153M+ disclosed", "hq": "San Diego, CA", "founded": "2019", "investors": "Ascenta Capital, Abingworth, NVIDIA, Illumina Ventures, Coatue, OrbiMed, Sequoia", "differentiator": "Weekly design-make-test cycles combine AI-generated molecular designs with automated experimental execution.", "entityType": "Company", "stage": "Clinical-stage platform", "region": "North America", "scienceDomains": [ "Drug discovery", "Chemistry", "Biology" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Enchant", "NeuralPLexer", "AI-driven discovery platform" ], "scores": { "autonomy": 3, "execution": 3, "verification": 3, "scientificEvidence": 3, "capital": 3, "traction": 3 }, "evidenceLevel": "Clinical pipeline", "accessModel": "Proprietary", "tractionSignals": [ "$100M Series B", "NVIDIA collaboration", "Weekly molecular design to biological data loops" ], "sourceIds": [ "official-iambic-platform", "fierce-iambic-series-b-202310" ] }, { "id": "causaly", "name": "Causaly", "tier": "tier3", "category": "Autonomous Researcher", "website": "https://www.causaly.com", "description": "Agentic AI platform for life-sciences R&D. Combines biomedical knowledge graphs, scientific retrieval, and governed agents for target identification, biomarker research, evidence review, and portfolio decisions.", "funding": "$93M+ disclosed", "hq": "London, UK; San Francisco, CA", "founded": "2018", "investors": "ICONIQ Growth, Index Ventures, Marathon Venture Capital, EBRD", "differentiator": "Science-grade agentic research layer used by pharma teams where factual grounding and traceable evidence matter more than generic chat.", "entityType": "Company", "stage": "Growth", "region": "Europe", "scienceDomains": [ "Biomedical research", "Drug discovery", "Evidence synthesis" ], "stackRoles": [ "Reason", "Verify", "Data" ], "products": [ "Causaly platform", "Agentic Research", "Biomedical knowledge graph" ], "scores": { "autonomy": 3, "execution": 0, "verification": 3, "scientificEvidence": 2, "capital": 3, "traction": 4 }, "evidenceLevel": "Commercial platform", "accessModel": "Proprietary", "tractionSignals": [ "$60M Series B", "Top pharma customers", "500M facts and 70M directional relationships claimed" ], "sourceIds": [ "official-causaly", "finsmes-causaly-series-b-202307" ] }, { "id": "iktos", "name": "Iktos", "tier": "tier3", "category": "Programmable Chemistry", "website": "https://iktos.ai", "description": "Generative AI and robotics platform for small-molecule drug discovery. Makya designs molecules, Spaya plans synthesis routes, and robotics execute synthesis, purification, analysis, and biological testing.", "funding": "EUR15.5M Series A plus EIC support", "hq": "Paris, France", "founded": "2016", "investors": "M Ventures, Debiopharm Innovation Fund, Omnes Capital, European Innovation Council", "differentiator": "Integrated design-make-test-analyze stack for medicinal chemistry rather than a design-only software platform.", "entityType": "Company", "stage": "Commercial platform", "region": "Europe", "scienceDomains": [ "Chemistry", "Drug discovery", "Medicinal chemistry" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Makya", "Spaya", "Iktos Robotics Chemistry", "Iktos Robotics Biology" ], "scores": { "autonomy": 3, "execution": 3, "verification": 2, "scientificEvidence": 3, "capital": 2, "traction": 3 }, "evidenceLevel": "Commercial platform", "accessModel": "Hybrid", "tractionSignals": [ "60+ collaborations", "2024 integrated discovery platform launch", "EIC Accelerator support" ], "sourceIds": [ "official-iktos", "official-iktos-about" ] }, { "id": "genesis-molecular-ai", "name": "Genesis Molecular AI", "tier": "tier3", "category": "AI-Native Biotech", "website": "https://www.genesis.ml", "description": "AI-first small-molecule drug discovery company formerly known as Genesis Therapeutics. GEMS combines generative AI, molecular simulation, and predictive models to design molecules for difficult targets.", "funding": "$280M+ disclosed", "hq": "Burlingame, CA", "founded": "2019", "investors": "Andreessen Horowitz, Fidelity, BlackRock, NVIDIA NVentures, T. Rowe Price, Menlo Ventures", "differentiator": "Stanford spinout using foundation-model and physics methods to iterate from virtual designs to experimental validation.", "entityType": "Company", "stage": "Private scaleup", "region": "North America", "scienceDomains": [ "Drug discovery", "Chemistry", "Molecular AI" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "GEMS", "Pearl", "AI-enabled small-molecule pipeline" ], "scores": { "autonomy": 3, "execution": 2, "verification": 3, "scientificEvidence": 3, "capital": 3, "traction": 3 }, "evidenceLevel": "Platform pipeline", "accessModel": "Proprietary", "tractionSignals": [ "$200M Series B", "Lilly and Genentech collaborations", "San Diego integrated lab" ], "sourceIds": [ "official-genesis-molecular-ai", "fierce-genesis-series-b-202308", "cen-genesis-series-b-202308" ] }, { "id": "argonne-national-laboratory", "name": "Argonne National Laboratory", "tier": "tier4", "category": "National Lab", "website": "https://www.anl.gov/autonomous-discovery", "description": "DOE national lab with an explicit Autonomous Discovery initiative. Uses AI, robotics, simulations, data platforms, and self-driving chemistry tools to automate experiments and scientific decision loops.", "funding": "Government", "hq": "Lemont, IL", "founded": "1946", "investors": "US Department of Energy", "differentiator": "Public autonomous-discovery program spanning robotic chemists, Polybot materials labs, AI-driven protein design, and autonomous data infrastructure.", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Materials", "Chemistry", "Biology", "Energy" ], "stackRoles": [ "Reason", "Execute", "Infrastructure", "Data", "Verify" ], "products": [ "Autonomous Discovery initiative", "Polybot", "Self-driving chemist", "Autonomous discovery data platform" ], "scores": { "autonomy": 4, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Research facility", "accessModel": "Public infrastructure", "tractionSignals": [ "Autonomous Discovery program", "Self-driving chemist project", "Polybot materials work" ], "sourceIds": [ "official-argonne-autonomous-discovery", "official-argonne-science101-autonomous-discovery" ] }, { "id": "oak-ridge-national-laboratory", "name": "Oak Ridge National Laboratory", "tier": "tier4", "category": "National Lab", "website": "https://www.ornl.gov/autonomousscience", "description": "DOE national lab building autonomous science platforms across instrumentation, computing, and laboratory automation. INTERSECT and the Autonomous Chemistry Lab connect AI, robotics, and HPC.", "funding": "Government", "hq": "Oak Ridge, TN", "founded": "1943", "investors": "US Department of Energy", "differentiator": "Combines leadership-class computing, user facilities, and self-driving labs under the Labs of the Future / Genesis Mission umbrella.", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Materials", "Chemistry", "Energy", "Computing" ], "stackRoles": [ "Reason", "Execute", "Infrastructure", "Data", "Verify" ], "products": [ "Autonomous Science program", "Autonomous Chemistry Lab", "INTERSECT initiative" ], "scores": { "autonomy": 4, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Research facility", "accessModel": "Public infrastructure", "tractionSignals": [ "Autonomous Science page", "Autonomous Chemistry Lab", "Genesis Mission role" ], "sourceIds": [ "official-ornl-autonomous-science", "official-ornl-autonomous-chemistry-lab" ] }, { "id": "lawrence-livermore-national-laboratory", "name": "Lawrence Livermore National Laboratory", "tier": "tier4", "category": "National Lab", "website": "https://www.llnl.gov", "description": "DOE/NNSA national lab applying AI, robotics, and high-performance computing to scientific discovery. APEX is a self-driving alloy-discovery platform for autonomous design, fabrication, and characterization.", "funding": "Government", "hq": "Livermore, CA", "founded": "1952", "investors": "US Department of Energy / NNSA", "differentiator": "APEX targets the full materials loop for 3D-printed alloys, from model-guided design through robotic sample preparation and testing.", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Materials", "Advanced manufacturing", "AI for science" ], "stackRoles": [ "Reason", "Execute", "Infrastructure", "Data", "Verify" ], "products": [ "APEX self-driving alloy lab", "AI/ML research programs", "Exascale computing infrastructure" ], "scores": { "autonomy": 3, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 2 }, "evidenceLevel": "Research facility", "accessModel": "Public infrastructure", "tractionSignals": [ "APEX platform", "Cornell collaboration", "Planned autonomous alloy workflow" ], "sourceIds": [ "official-llnl-apex-202507", "str-llnl-apex-202601" ] }, { "id": "nist", "name": "NIST", "tier": "tier4", "category": "National Lab", "website": "https://www.nist.gov/autonomous-laboratories", "description": "US standards and measurement institute with autonomous-laboratory programs in materials, formulations, metrology, and biofoundry automation. Focuses on closed-loop experimentation and interoperability.", "funding": "Government", "hq": "Gaithersburg, MD", "founded": "1901", "investors": "US Department of Commerce", "differentiator": "Standards-oriented role for autonomous labs, including CAMEO, Autonomous Formulation Lab, Hermes software, and measurement-focused self-driving systems.", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Materials", "Metrology", "Formulations", "Biology" ], "stackRoles": [ "Execute", "Verify", "Infrastructure", "Data" ], "products": [ "Autonomous laboratories program", "Autonomous Formulation Lab", "CAMEO", "Hermes" ], "scores": { "autonomy": 3, "execution": 4, "verification": 4, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Research facility", "accessModel": "Public infrastructure", "tractionSignals": [ "Autonomous laboratories portfolio", "NIST SP 1320", "Autonomous Formulation Lab" ], "sourceIds": [ "official-nist-autonomous-labs", "nist-autonomous-systems-materials-202009", "nist-autonomous-formulation-lab-202508" ] }, { "id": "nrel", "name": "NREL", "tier": "tier4", "category": "National Lab", "website": "https://www.nrel.gov/materials-science/autonomous-experimentation", "description": "DOE national lab with autonomous experimentation capabilities for energy materials. Work includes autonomous synthesis, autonomous characterization, and AI-assisted control software for instruments.", "funding": "Government", "hq": "Golden, CO", "founded": "1977", "investors": "US Department of Energy", "differentiator": "Energy-materials focus: autonomous sputter deposition, autonomous microscopy, and open control-code workflows for real-world materials instruments.", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Materials", "Energy", "Characterization" ], "stackRoles": [ "Execute", "Data", "Verify", "Infrastructure" ], "products": [ "Autonomous experimentation program", "Autonomous synthesis", "Autonomous characterization", "AI-generated control software" ], "scores": { "autonomy": 3, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Research facility", "accessModel": "Public infrastructure", "tractionSignals": [ "Autonomous sputter synthesis", "Autonomous characterization", "Digital Discovery control-software paper" ], "sourceIds": [ "official-nrel-autonomous-experimentation", "rsc-text-to-test-202410" ] }, { "id": "brookhaven-national-laboratory", "name": "Brookhaven National Laboratory", "tier": "tier4", "category": "National Lab", "website": "https://www.bnl.gov", "description": "DOE national lab and user-facility operator applying AI, robotics, and autonomous experiment steering to materials and photon-science workflows at NSLS-II and CFN.", "funding": "Government", "hq": "Upton, NY", "founded": "1947", "investors": "US Department of Energy", "differentiator": "Large-scale user-facility angle: autonomous steering, Bluesky/robotics infrastructure, and AI-driven nanostructure discovery at Brookhaven facilities.", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Materials", "Photon science", "Nanoscience" ], "stackRoles": [ "Execute", "Data", "Verify", "Infrastructure" ], "products": [ "NSLS-II autonomous steering tools", "CFN autonomous materials discovery", "Robotics and Bluesky workflows" ], "scores": { "autonomy": 3, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Research facility", "accessModel": "Public infrastructure", "tractionSignals": [ "NSLS-II DSSI robotics work", "AI-discovered nanostructures", "ILLUMINE autonomous experiment steering" ], "sourceIds": [ "official-bnl-nsls2-dssi", "sciencedaily-bnl-ai-nanostructures-202301" ] }, { "id": "materials-genome-initiative", "name": "Materials Genome Initiative", "tier": "tier4", "category": "Academic Network", "website": "https://www.mgi.gov", "description": "US federal multi-agency initiative for accelerated materials discovery and deployment. Coordinates infrastructure, data, autonomous experimentation, and policy for materials innovation.", "funding": "Government program", "hq": "United States", "founded": "2011", "investors": "US federal agencies", "differentiator": "Foundational policy and coordination layer behind US materials-acceleration infrastructure, including autonomous experimentation roadmapping.", "entityType": "Government initiative", "stage": "Policy and research network", "region": "North America", "scienceDomains": [ "Materials", "Manufacturing", "Policy", "Data infrastructure" ], "stackRoles": [ "Infrastructure", "Data", "Verify" ], "products": [ "Materials Genome Initiative", "Autonomous experimentation reports", "Federal agency coordination" ], "scores": { "autonomy": 1, "execution": 1, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 4 }, "evidenceLevel": "Government initiative", "accessModel": "Public infrastructure", "tractionSignals": [ "2011 launch", "Multi-agency federal coordination", "Autonomous experimentation workshop report" ], "sourceIds": [ "official-mgi-about", "official-mgi-autonomous-experimentation-report" ] }, { "id": "chai-discovery", "name": "Chai Discovery", "tier": "tier2", "category": "AI-Native Biotech", "website": "https://www.chaidiscovery.com", "description": "AI drug and biologics design company building a computer-aided design suite for molecules, with Chai-2 focused on zero-shot antibody and protein-binder design.", "funding": "$225M+", "valuation": "$1.3B", "hq": "San Francisco, CA", "founded": "2024", "investors": "Oak HC/FT, General Catalyst, Menlo Ventures, OpenAI, Thrive Capital, Dimension", "differentiator": "Chai-2 reports double-digit experimental success rates for de novo antibody design, a 100-fold improvement over previous computational methods, and a Lilly collaboration using proprietary Lilly data.", "entityType": "Company", "stage": "Growth", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery", "Protein design" ], "stackRoles": [ "Reason", "Data", "Verify" ], "products": [ "Chai-1", "Chai-2", "Molecular design platform" ], "scores": { "autonomy": 3, "execution": 1, "verification": 3, "scientificEvidence": 4, "capital": 4, "traction": 4 }, "evidenceLevel": "Wet-lab validated model", "accessModel": "Commercial partnership", "tractionSignals": [ "$130M Series B at $1.3B valuation", "$225M+ total funding", "Eli Lilly biologics collaboration" ], "sourceIds": [ "official-chai-discovery", "businesswire-chai-series-b-202512", "biospace-chai-lilly-202601", "biorxiv-chai2-zero-shot-202507" ] }, { "id": "converge-bio", "name": "Converge Bio", "tier": "tier3", "category": "AI-Native Biotech", "website": "https://converge-bio.com", "description": "Generative AI platform for drug discovery and development, spanning target discovery, antibody design, and protein manufacturing optimization.", "funding": "$30M", "hq": "Boston, MA / Tel Aviv, Israel", "founded": "2024", "investors": "Bessemer Venture Partners, TLV Partners, Vintage Investment Partners, Saras Capital", "differentiator": "Over a dozen pharma and biotech customers and 40+ completed programs; models plug into existing workflows and are trained on large-scale proprietary and curated public datasets.", "entityType": "Company", "stage": "Series A", "region": "North America / Middle East", "scienceDomains": [ "Biology", "Drug discovery", "Protein engineering" ], "stackRoles": [ "Reason", "Data", "Verify" ], "products": [ "Target discovery models", "Antibody design system", "Protein manufacturing optimization system" ], "scores": { "autonomy": 2, "execution": 1, "verification": 2, "scientificEvidence": 2, "capital": 3, "traction": 3 }, "evidenceLevel": "Commercial traction", "accessModel": "Commercial partnership", "tractionSignals": [ "$25M Series A", "$30M total raised", "40+ pharma and biotech programs" ], "sourceIds": [ "official-converge-series-a-202601", "techcrunch-converge-series-a-202601" ] }, { "id": "generare", "name": "Generare", "tier": "tier3", "category": "AI-Native Natural Products", "website": "https://generare.bio", "description": "Techbio company decoding microbial genomes to express silent chemistry and generate proprietary novel small-molecule data for drug discovery.", "funding": "€25M", "hq": "Paris, France", "founded": "2023", "investors": "Alven, Daphni, Galion.exe, Teampact Ventures, VIVES Partners", "differentiator": "Reported 200 previously uncharacterized small molecules in 2025 and plans to scale its proprietary molecular dataset ten-fold by 2027.", "entityType": "Company", "stage": "Series A", "region": "Europe", "scienceDomains": [ "Natural products", "Drug discovery", "Synthetic biology" ], "stackRoles": [ "Data", "Execute", "Reason" ], "products": [ "Microbial genome-screening platform", "Cryptic small-molecule dataset", "Synthetic-biology expression workflow" ], "scores": { "autonomy": 2, "execution": 3, "verification": 2, "scientificEvidence": 3, "capital": 3, "traction": 3 }, "evidenceLevel": "Commercial data-generation platform", "accessModel": "Commercial partnership", "tractionSignals": [ "€20M Series A", "200 previously uncharacterized small molecules", "10x dataset scale target by 2027" ], "sourceIds": [ "alven-generare-series-a-202604", "biologydigital-generare-series-a-202604" ] }, { "id": "alloy-therapeutics", "name": "Alloy Therapeutics", "tier": "tier4", "category": "Biotech Infrastructure", "website": "https://www.alloytx.com", "description": "Tech-enabled biotech infrastructure and integrated services platform for biologics discovery and development, combining AI, real-world data, and wet-lab execution.", "funding": "$40M Series E", "valuation": "$1B", "hq": "Boston, MA", "founded": "2017", "investors": "8VC, JIC Venture Growth Investments, Mubadala Capital, Founders Fund, Alexandria Venture Investments", "differentiator": "200+ partners, 100+ licensed therapeutic programs, and 22 clinical programs; included as scope-review biotech infrastructure rather than an autonomous-lab operator.", "entityType": "Company", "stage": "Growth", "region": "North America", "scienceDomains": [ "Biologics", "Drug discovery", "Biotech infrastructure" ], "stackRoles": [ "Execute", "Data", "Infrastructure" ], "products": [ "Biologics discovery platforms", "Integrated preclinical services", "AI and real-world-data infrastructure" ], "scores": { "autonomy": 1, "execution": 3, "verification": 2, "scientificEvidence": 2, "capital": 3, "traction": 4 }, "evidenceLevel": "Commercial infrastructure", "accessModel": "Commercial partnership", "tractionSignals": [ "$40M Series E at $1B valuation", "200+ partners", "22 clinical programs" ], "sourceIds": [ "businesswire-alloy-series-e-202604", "citybiz-alloy-series-e-202604" ] }, { "id": "robochem-flex", "name": "RoboChem-Flex", "tier": "tier4", "category": "Academic Self-Driving Lab", "website": "https://www.nature.com/articles/s44160-026-01053-0", "description": "Academic low-cost modular self-driving laboratory for automated chemical reaction optimization, with Python control software and Bayesian optimization.", "funding": "Academic research", "hq": "Netherlands", "founded": "2026", "investors": "University research program", "differentiator": "Affordable SDL infrastructure validated across six case studies with closed-loop and human-in-the-loop optimization modes.", "entityType": "Academic platform", "stage": "Peer-reviewed prototype", "region": "Europe", "scienceDomains": [ "Chemistry", "Lab automation", "Optimization" ], "stackRoles": [ "Execute", "Reason", "Infrastructure", "Verify" ], "products": [ "RoboChem-Flex modular SDL", "Python control framework", "Bayesian optimization workflow" ], "scores": { "autonomy": 3, "execution": 3, "verification": 3, "scientificEvidence": 4, "capital": 1, "traction": 2 }, "evidenceLevel": "Peer-reviewed prototype", "accessModel": "Open research", "tractionSignals": [ "Nature Synthesis 2026", "Six validation case studies", "Low-cost modular hardware" ], "sourceIds": [ "nature-robochem-flex-202604" ] }, { "id": "scivity-labs", "name": "Scivity Labs", "tier": "tier3", "category": "Autonomous Science Platform", "website": "https://scivity.org", "description": "Autonomous science platform focused on computational research workflows that design, execute, and verify scientific work end-to-end, with public emphasis on validation outcomes and evidence surfaces.", "funding": "Pre-seed, undisclosed", "hq": "Yerevan, Armenia", "founded": "2025", "investors": "Undisclosed", "differentiator": "Verification-first positioning: public outputs and evidence surfaces are open, while thresholds, calibration, and check-composition logic remain private.", "entityType": "Company", "stage": "Pre-seed", "region": "Europe", "scienceDomains": [ "ML research", "Biology", "Chemistry", "General science" ], "stackRoles": [ "Reason", "Execute", "Verify", "Infrastructure" ], "products": [ "Autonomous Laboratory Platform", "Verification surface", "AI for Science Landscape" ], "scores": { "autonomy": 3, "execution": 2, "verification": 2, "scientificEvidence": 1, "capital": 1, "traction": 1 }, "scoreNotes": { "verification": "Conservative score: public materials describe validation outcomes while thresholds, calibration, and check-composition logic remain private." }, "evidenceLevel": "Self-reported platform", "accessModel": "Proprietary", "tractionSignals": [ "Pre-seed company", "Public verification surface", "First autonomous end-to-end research run disclosed" ], "sourceIds": [ "official-scivity-home-202605", "official-scivity-verification-202603", "official-scivity-changelog-202604" ] }, { "id": "tetsuwan-scientific", "name": "Tetsuwan Scientific", "tier": "tier2", "category": "Self-Driving Lab Software", "website": "https://tetsuwan.com", "description": "Execution-layer startup building ResearchOS and a future cloud lab so researchers or AI agents can translate natural-language procedures into machine-ready wet-lab workflows.", "funding": "$2.7M pre-seed", "hq": "San Francisco, CA", "founded": "2023", "investors": "2048 Ventures, Carbon Silicon Ventures, Referent Ventures, Everywhere Ventures", "differentiator": "ResearchOS positions itself as the interface between researchers, agents, and automated biology labs, with cloud-lab ambitions from day one.", "entityType": "Company", "stage": "Pre-seed", "region": "North America", "scienceDomains": [ "Biology", "Lab automation", "Cloud lab" ], "stackRoles": [ "Reason", "Execute", "Infrastructure" ], "products": [ "ResearchOS", "Tetsuwan Cloud Lab", "LabAutoWiki" ], "scores": { "autonomy": 2, "execution": 3, "verification": 1, "scientificEvidence": 1, "capital": 1, "traction": 2 }, "evidenceLevel": "Pilot", "accessModel": "Proprietary", "tractionSignals": [ "$2.7M pre-seed", "First lab deployment reported", "ResearchOS execution-layer positioning" ], "sourceIds": [ "official-tetsuwan-home-2026", "techcrunch-tetsuwan-preseed-202412", "synbiobeta-tetsuwan-preseed-202411" ] }, { "id": "microsoft-discovery", "name": "Microsoft Discovery", "tier": "tier5", "category": "Enterprise R&D Platform", "website": "https://azure.microsoft.com/en-us/solutions/discovery", "description": "Microsoft's enterprise agentic R&D platform for hypothesis generation, simulation, analysis, and iterative experimentation across life sciences, materials, semiconductors, and engineering workflows.", "funding": "Microsoft-funded preview", "hq": "Redmond, WA", "founded": "2025", "investors": "Microsoft", "differentiator": "Pairs a graph-based Discovery Engine with Azure governance, HPC, and partner integrations to move from scientific reasoning to governed enterprise execution.", "entityType": "Platform initiative", "stage": "Preview", "region": "North America", "scienceDomains": [ "Biology", "Chemistry", "Materials", "Semiconductors" ], "stackRoles": [ "Reason", "Execute", "Verify", "Data", "Infrastructure" ], "products": [ "Microsoft Discovery Engine", "Azure-governed R&D platform", "Agentic experiment loop" ], "scores": { "autonomy": 3, "execution": 2, "verification": 3, "scientificEvidence": 2, "capital": 4, "traction": 2 }, "evidenceLevel": "Enterprise preview", "accessModel": "Preview / enterprise", "tractionSignals": [ "Build 2025 launch", "Expanded preview in 2026", "Azure-native governance and HPC" ], "sourceIds": [ "official-microsoft-discovery-solution-202604", "official-microsoft-discovery-intro-202505", "official-microsoft-discovery-preview-202604", "techcrunch-microsoft-discovery-202505" ] }, { "id": "anthropic", "name": "Anthropic", "tier": "tier5", "category": "Frontier AI Lab", "website": "https://www.anthropic.com", "description": "Frontier model developer (Claude family). October 2025 launched Claude for Life Sciences with the Allen Institute and HHMI as founding scientific partners and connectors to Benchling, PubMed, 10x Genomics, ClinicalTrials.gov, and Synapse. April 2026 acquired Coefficient Bio (all-stock, ~$400M) as its first M&A into a new domain. Top of the Ai2 AstaBench scientific-agent leaderboard (Claude Opus 4.7 at 58.0%).", "funding": "$10B+ raised across rounds", "valuation": "$183B (reported 2025)", "hq": "San Francisco, CA", "founded": "2021", "investors": "Amazon, Google, Lightspeed, Spark Capital, Salesforce, Menlo Ventures", "differentiator": "Best-in-class scientific reasoning on AstaBench paired with the first dedicated frontier-lab life-sciences product and a Coefficient Bio acqui-hire to anchor the science team.", "entityType": "Frontier AI lab", "stage": "Frontier platform", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery", "General science", "Reasoning" ], "stackRoles": [ "Reason", "Verify", "Data" ], "products": [ "Claude family", "Claude for Life Sciences", "BioMysteryBench", "Agent Skills", "Coefficient Bio team (acquired)" ], "scores": { "autonomy": 3, "execution": 1, "verification": 3, "scientificEvidence": 3, "capital": 4, "traction": 4 }, "evidenceLevel": "Frontier platform", "accessModel": "Proprietary", "tractionSignals": [ "Top of AstaBench scientific-agent leaderboard", "Allen Institute and HHMI founding partnerships", "Coefficient Bio acquisition", "Claude for Life Sciences launch" ], "sourceIds": [ "official-anthropic-claude-life-sciences-202510", "official-anthropic-allen-hhmi-202510", "official-anthropic-biomysterybench-202604", "techcrunch-anthropic-coefficient-bio-202604", "biospace-anthropic-coefficient-202604", "ai2-astabench-update-202604" ] }, { "id": "ai2", "name": "Ai2", "tier": "tier2", "category": "Autonomous Researcher", "website": "https://allenai.org", "description": "Allen Institute for AI — nonprofit research institute founded by Paul Allen. In August 2025 launched Asta, an open agentic ecosystem for science covering scholarly research assistants, the AstaBench benchmark suite (2,400+ problems across 11 benchmarks), and a developer toolkit with open-source agents, post-trained science LMs, and a 200M-paper Scientific Corpus Tool. AstaBench accepted as ICLR 2026 oral.", "funding": "Endowment-backed nonprofit", "hq": "Seattle, WA", "founded": "2014", "investors": "Paul G. Allen Family Foundation (founding)", "differentiator": "Open scientific-agent stack and the de-facto benchmark for scientific AI agents, already adopted by UK AISI, General Reasoning, Elicit, SciSpace, Distyl AI, and EvoScientist.", "entityType": "Nonprofit research institute", "stage": "Research platform", "region": "North America", "scienceDomains": [ "General science", "Literature", "Biology", "Reasoning" ], "stackRoles": [ "Reason", "Verify", "Data" ], "products": [ "Asta agents", "AstaBench", "Scientific Corpus Tool", "OLMo models", "Semantic Scholar" ], "scores": { "autonomy": 3, "execution": 1, "verification": 4, "scientificEvidence": 4, "capital": 3, "traction": 3 }, "evidenceLevel": "Open research", "accessModel": "Open", "tractionSignals": [ "AstaBench ICLR 2026 oral", "UK AISI and General Reasoning adoption", "Submissions from Elicit, SciSpace, Distyl AI, EvoScientist", "Asta runs on 108M+ abstracts and 200M+ paper corpus" ], "sourceIds": [ "official-ai2-asta-launch-202508", "official-ai2-asta-platform", "official-astabench", "ai2-astabench-blog-202508", "ai2-astabench-update-202604", "businesswire-ai2-asta-202508" ] }, { "id": "amazon-bio-discovery", "name": "Amazon Bio Discovery", "tier": "tier5", "category": "Enterprise R&D Platform", "website": "https://aws.amazon.com/health/biology-foundation-models/", "description": "AWS's agentic AI application for drug research, announced April 28, 2026. Exposes 40+ biology foundation models, supports custom uploads and fine-tuning, and integrates with CRO and cloud-lab partners (Ginkgo Nebula, Twist Bioscience, with A-Alpha Bio anticipated) so that agents select BioFMs, route candidates to wet-lab partners for synthesis and testing, and return experimental results into the loop on Bedrock.", "funding": "Amazon-funded service", "hq": "Seattle, WA", "founded": "2026", "investors": "Amazon", "differentiator": "First Big Tech lab-in-the-loop platform that closes the agent → BioFM → wet-lab → result cycle directly on AWS infrastructure, with an early Memorial Sloan Kettering project compressing a year of antibody triage into weeks.", "entityType": "Platform initiative", "stage": "Preview", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery" ], "stackRoles": [ "Reason", "Execute", "Data", "Infrastructure" ], "products": [ "Amazon Bio Discovery", "Biology foundation model catalog", "Lab-in-the-loop orchestration", "Bedrock agents" ], "scores": { "autonomy": 3, "execution": 2, "verification": 2, "scientificEvidence": 2, "capital": 4, "traction": 2 }, "evidenceLevel": "Enterprise preview", "accessModel": "Preview / enterprise", "tractionSignals": [ "April 2026 launch with 40+ BioFMs", "Ginkgo Nebula and Twist Bioscience integration", "Memorial Sloan Kettering early project", "Closed lab-in-the-loop on AWS" ], "sourceIds": [ "official-aws-bio-discovery-blog-202604", "genengnews-aws-bio-discovery-202604", "aipedia-aws-bio-discovery-202604", "official-ginkgo-q1-2026" ] }, { "id": "profluent", "name": "Profluent", "tier": "tier2", "category": "AI-Native Biotech", "website": "https://www.profluent.bio", "description": "Programmable-biology company using frontier protein language models. November 2025 closed a $106M Series B co-led by Altimeter Capital and Bezos Expeditions, taking total funding to $150M. May 2026 announced a strategic collaboration with Eli Lilly for custom site-specific recombinases addressing severe-unmet-need diseases, with up to $2.25B in development and commercial milestones plus tiered royalties. Released OpenCRISPR-1, the first AI-generated CRISPR gene editor.", "funding": "$150M total ($106M Series B Nov 2025)", "hq": "Berkeley, CA", "founded": "2022", "investors": "Altimeter Capital, Bezos Expeditions, Spark Capital, Insight Partners, Air Street Capital", "differentiator": "Demonstrated scaling laws for protein design (NeurIPS 2025 spotlight) and the first wholly AI-designed CRISPR system (Nature 2025), then converted that into a $2.25B Lilly recombinase pact in May 2026.", "entityType": "Company", "stage": "Growth", "region": "North America", "scienceDomains": [ "Biology", "Protein design", "Gene editing" ], "stackRoles": [ "Reason", "Data", "Verify" ], "products": [ "Protein foundation models", "OpenCRISPR-1", "ProGen", "Site-specific recombinase platform" ], "scores": { "autonomy": 2, "execution": 2, "verification": 3, "scientificEvidence": 4, "capital": 4, "traction": 4 }, "evidenceLevel": "Wet-lab validated model", "accessModel": "Hybrid", "tractionSignals": [ "$106M Series B led by Altimeter and Bezos", "$2.25B Lilly recombinase pact", "OpenCRISPR-1 open release", "Nature 2025 AI-designed CRISPR system" ], "sourceIds": [ "official-profluent", "businesswire-profluent-series-b-202511", "biospace-profluent-lilly-202604", "airstreet-profluent-lilly-202604" ] } ], "companies": [ { "id": "periodic-labs", "name": "Periodic Labs", "tier": "tier1", "category": "Full-Stack Lab", "website": "https://periodic.com", "description": "Autonomous physical labs for materials and chemistry. AI scientists generate hypotheses; robotic labs run synthesis and characterization; results feed back as training data unavailable anywhere else.", "funding": "$300M seed", "valuation": "$1.5B", "hq": "San Francisco, CA", "founded": "2025", "investors": "Andreessen Horowitz, Nvidia, Bezos, Eric Schmidt, DST, Accel, Jeff Dean, Elad Gil", "differentiator": "Ex-OpenAI (Liam Fedus, co-creator of ChatGPT) and ex-DeepMind (Ekin Dogus Cubuk, GNoME). Building superconductors as first target." }, { "id": "lila-sciences", "name": "Lila Sciences", "tier": "tier1", "category": "Full-Stack Lab", "website": "https://lila.ai", "description": "AI Science Factories — integrated reasoning models, robotic labs, and verifiers running the scientific method across life, chemistry, and materials sciences.", "funding": "$550M", "valuation": "$1.3B", "hq": "Cambridge, MA", "founded": "2023", "investors": "Flagship Pioneering, Braidwell, Collective Global, General Catalyst, March, ARK, NVIDIA NVentures, ADIA", "differentiator": "Born out of Flagship Pioneering. George Church as chief scientist. Facilities in Boston, SF, London." }, { "id": "sakana-ai", "name": "Sakana AI", "tier": "tier1", "category": "Autonomous Researcher", "website": "https://sakana.ai", "description": "The AI Scientist — autonomous ML research agent. Produces end-to-end research papers with minimal human involvement.", "funding": "$379M cumulative", "valuation": "$2.65B", "hq": "Tokyo, Japan", "founded": "2023", "investors": "NEA, Khosla, Lux Capital, NVIDIA, MUFG, In-Q-Tel", "differentiator": "Japan's most valuable private AI startup. Co-founded by David Ha and Llion Jones (Transformer paper)." }, { "id": "futurehouse", "name": "FutureHouse", "tier": "tier1", "category": "Autonomous Researcher", "website": "https://futurehouse.org", "description": "Nonprofit AI research lab building autonomous scientists for biology. Released the Crow, Falcon, Owl, and Phoenix agent family, added ether0 and DISCO in 2026, and spun out Edison Scientific in late 2025.", "funding": "~$20M grants + $70M spinout seed", "hq": "San Francisco, CA", "founded": "2023", "investors": "Eric Schmidt, Schmidt Futures; Edison spinout led by Spark Capital", "differentiator": "Biology-first multi-agent system spanning literature, reasoning, and wet-lab-linked model work. DISCO and other 2026 releases show the nonprofit still moving the research frontier while Edison commercializes the stack." }, { "id": "autoscience-institute", "name": "Autoscience Institute", "tier": "tier1", "category": "Autonomous Researcher", "website": "https://autoscience.ai", "description": "Autonomous agents for ML research. Carl ideates and writes papers; Mira embeds new research directly into production ML models.", "funding": "$14M seed", "hq": "San Francisco, CA", "founded": "2023", "investors": "General Catalyst, Toyota Ventures, Perplexity Fund, S32", "differentiator": "Carl was the first AI system to generate a full-length peer-reviewed workshop paper (ICLR 2025, 3 of 4 submissions accepted)." }, { "id": "thesis", "name": "Thesis", "tier": "tier1", "category": "Autonomous Researcher", "website": "https://www.ycombinator.com/companies/thesis", "description": "Applied research lab automating parts of the ML research workflow — experiment planning, training, evaluation, iteration.", "funding": "~$3M", "hq": "San Francisco, CA", "founded": "2025", "investors": "Y Combinator (F25 batch)", "differentiator": "State-of-the-art on OpenAI MLE-Bench. Founded by Stanford AI Lab alumni working with Chelsea Finn and Andrew Ng." }, { "id": "zeon-systems", "name": "Zeon Systems", "tier": "tier1", "category": "Lab Robotics", "website": "https://zeonsystems.ai", "description": "Natural-language-to-robot-arm layer for lab automation. Scientists type protocols in English; software programs off-the-shelf robotics to execute them.", "funding": "$0.5M", "hq": "San Francisco, CA", "founded": "2025", "investors": "Y Combinator (X25), FCVC, A* Capital", "differentiator": "Pilots at UCSF and Stanford. Computer-vision labware detection. Positions itself as the 'action endpoint' for AI science agents." }, { "id": "automata", "name": "Automata", "tier": "tier2", "category": "Lab Robotics", "website": "https://automata.tech", "description": "Lab-automation operating system for AI-ready wet labs. Modular robotics, orchestration software, and unified data infrastructure to turn physical labs into programmable, repeatable systems.", "funding": "$45M Series C", "hq": "London, UK", "founded": "2015", "investors": "Dimension (lead), Danaher Ventures, Tru Arrow Partners, Octopus Ventures, Entrepreneurs First", "differentiator": "Pharma-first go-to-market — five of the top pharma companies are customers. Danaher strategic partnership on the Series C, with a board seat." }, { "id": "cenevo", "name": "Cenevo", "tier": "tier2", "category": "Lab Orchestration / Lab Informatics", "website": "https://www.cenevo.com", "description": "Lab orchestration and informatics platform formed from Labguru and Titian Software. Provides ELN/LIMS, sample management, workflow automation, protocol conversion, and AI agents for regulated labs.", "funding": "Private", "hq": "United States / United Kingdom", "founded": "2025 rebrand", "investors": "Battery Ventures", "differentiator": "Commercial informatics footprint with 950+ customer organizations, 45,000+ users, and 8 of the top 10 pharma companies, plus agentic orchestration features with approvals, audit trails, and GxP positioning." }, { "id": "emerald-cloud-lab", "name": "Emerald Cloud Lab", "tier": "tier2", "category": "Cloud Lab", "website": "https://www.emeraldcloudlab.com", "description": "Remote-controlled cloud lab — programmable chemistry and biology infrastructure accessed via a full scientific-computing environment.", "funding": "Private, undisclosed", "hq": "South San Francisco, CA", "founded": "2010", "investors": "Private", "differentiator": "Most mature programmable cloud lab. Used by pharma, Emerald Therapeutics, and academic groups at CMU." }, { "id": "strateos", "name": "Strateos", "tier": "tier2", "category": "Cloud Lab", "website": "https://strateos.com", "description": "Robotic cloud lab for life sciences. Automated workflows for drug discovery, synthesis, and assays, accessible via API.", "funding": "$56M", "hq": "San Diego, CA", "founded": "2012", "investors": "Google Ventures, Mubadala", "differentiator": "Pharmaceutical R&D focus. API-first design for remote experiment orchestration." }, { "id": "chemify", "name": "Chemify", "tier": "tier2", "category": "Programmable Chemistry", "website": "https://chemify.io", "description": "Chemputer platform — universal chemical synthesis via programmable hardware and AI-driven reaction planning. Commercialization of Lee Cronin's Glasgow work.", "funding": "£43M", "hq": "Glasgow, UK", "founded": "2023", "investors": "Triatomic Capital, DCVC", "differentiator": "Universal programmable synthesis. Chemistry-as-code academic pedigree (Cronin lab)." }, { "id": "recursion-pharmaceuticals", "name": "Recursion Pharmaceuticals", "tier": "tier3", "category": "AI-Native Biotech", "website": "https://www.recursion.com", "description": "Publicly traded AI-native drug discovery platform. Industrial-scale phenomics, ML-driven target identification, and image-based screening at PB-scale.", "funding": "Public (NASDAQ: RXRX)", "valuation": "~$2.5B", "hq": "Salt Lake City, UT", "founded": "2013", "investors": "Public market", "differentiator": "Largest industrial AI biology platform. Partnerships with Bayer, Roche, Nvidia. Merged with Exscientia in 2024." }, { "id": "atinary", "name": "Atinary", "tier": "tier3", "category": "Self-Driving Lab Software", "website": "https://atinary.com", "description": "Self-driving lab orchestration and execution layer. Runs SDLabs software plus physical Scientific Discovery Factories for chemistry, materials, and pharma R&D teams.", "funding": "Undisclosed", "hq": "Lausanne, Switzerland", "founded": "2019", "investors": "Private", "differentiator": "Commercial hardware-and-software self-driving lab stack that connects no-code AI optimization with robotics and analytical instruments." }, { "id": "ginkgo-bioworks", "name": "Ginkgo Bioworks", "tier": "tier3", "category": "Platform Biotech", "website": "https://www.ginkgobioworks.com", "description": "Synthetic biology foundry. Engineered organisms as a service; increasingly AI-augmented across the design-build-test-learn cycle.", "funding": "Public (NYSE: DNA)", "valuation": "~$1B", "hq": "Boston, MA", "founded": "2008", "investors": "Public market", "differentiator": "Largest organism engineering platform. Scale advantage in biological data generation." }, { "id": "radical-ai", "name": "Radical AI", "tier": "tier3", "category": "AI-Native Materials", "website": "https://radicalai.com", "description": "AI platform for materials discovery. Focus on industrial biomanufacturing and novel functional materials.", "funding": "$13.5M", "hq": "New York, NY", "founded": "2022", "investors": "Plural, Y Combinator", "differentiator": "Materials-focused foundation models. Commercialization via industrial partnerships." }, { "id": "xaira-therapeutics", "name": "Xaira Therapeutics", "tier": "tier3", "category": "AI-Native Biotech", "website": "https://www.xaira.com", "description": "AI-driven drug discovery and design company. Uses ML from hypothesis through clinical development.", "funding": "$1B+ Series A", "hq": "South San Francisco, CA", "founded": "2024", "investors": "ARCH Venture Partners, Foresite Labs, F-Prime, NEA, Sequoia, Lux, Lightspeed", "differentiator": "Largest initial biotech commitment in ARCH history. Co-founded by Nobel laureate David Baker and Marc Tessier-Lavigne." }, { "id": "helical", "name": "Helical", "tier": "tier3", "category": "AI-Native Biotech", "website": "https://www.helical.bio", "description": "Virtual AI lab for pharma R&D. Converts biological foundation models into reproducible in-silico discovery workflows for target identification, biomarker discovery, and therapeutic design.", "funding": "$10M seed", "valuation": "undisclosed", "hq": "London, UK", "founded": "2024", "investors": "redalpine, Gradient, BoxGroup, Frst, Aidan Gomez, Clement Delangue", "differentiator": "Production deployments with top-20 pharma teams, including a public Pfizer biomarker collaboration; bridges biologists and ML engineers through shared virtual-lab and model-factory surfaces." }, { "id": "fathom-therapeutics", "name": "Fathom Therapeutics", "tier": "tier3", "category": "AI-Native Biotech", "website": "https://www.fathomtx.com", "description": "Physics- and AI-driven small-molecule drug design company formerly known as Atommap. Microcosmos simulates protein motion and interactions at atomic resolution to guide lab-in-the-loop discovery.", "funding": "$47M Series A", "valuation": "undisclosed", "hq": "New York, NY", "founded": "2023", "investors": "Sutter Hill Ventures, Chemistry, Alexandria Venture Investments, Empire State Development's NY Ventures", "differentiator": "Founded by Huafeng Xu, Yujie Wu, and Jesus Izaguirre; combines quantum chemistry, molecular dynamics, and AI to model dynamic protein behavior rather than static structures." }, { "id": "10x-science", "name": "10x Science", "tier": "tier4", "category": "AI-Native Biotech", "website": "https://10xscience.com", "description": "AI-native platform for next-generation protein characterization. Automates mass-spectrometry interpretation so biologic drug candidates can be characterized and quality-assessed faster.", "funding": "$4.8M seed", "valuation": "undisclosed", "hq": "San Francisco, CA", "founded": "2025", "investors": "Initialized Capital, Y Combinator, Civilization Ventures, Founder Factor", "differentiator": "Stanford/Bertozzi-lab founding team attacking the verification bottleneck created by AI-generated biologics candidates." }, { "id": "a-lab-lbnl", "name": "A-Lab (LBNL)", "tier": "tier4", "category": "National Lab Facility", "website": "https://a-lab.lbl.gov", "description": "Autonomous materials synthesis lab at Lawrence Berkeley National Lab. AI-driven synthesis of novel inorganic materials.", "funding": "Government", "hq": "Berkeley, CA", "founded": "2022", "investors": "US Department of Energy", "differentiator": "Reference implementation for autonomous materials labs. Nature-published 2023; subject of a major 2024 critique and 2026 correction — the defining verification case study of the field." }, { "id": "acceleration-consortium", "name": "Acceleration Consortium", "tier": "tier4", "category": "Academic Network", "website": "https://acceleration.utoronto.ca", "description": "International consortium based at University of Toronto. Coordinates self-driving labs across member institutions; publishes protocols and standards.", "funding": "$200M+ (CFREF grant)", "hq": "Toronto, Canada", "founded": "2021", "investors": "Canada First Research Excellence Fund", "differentiator": "Standards body for the autonomous-lab field. Over 30 member institutions worldwide." }, { "id": "pnnl", "name": "PNNL", "tier": "tier4", "category": "National Lab", "website": "https://www.pnnl.gov", "description": "Pacific Northwest National Laboratory. DOE lab with autonomous experimentation programs in chemistry and energy. Part of the DOE × DeepMind Genesis Mission network.", "funding": "Government", "hq": "Richland, WA", "founded": "1965", "investors": "US Department of Energy", "differentiator": "National-lab infrastructure. Deep expertise in autonomous chemistry and energy systems." }, { "id": "google-deepmind", "name": "Google DeepMind", "tier": "tier5", "category": "Frontier AI Lab", "website": "https://deepmind.google", "description": "AlphaFold, AlphaProteo, GNoME, AlphaGeometry, AlphaEvolve — state-of-the-art scientific AI, without a dedicated autonomous-lab product line.", "funding": "Alphabet subsidiary", "hq": "London, UK", "founded": "2010", "investors": "Alphabet", "differentiator": "Deepest scientific AI bench in the industry. Genesis Mission partnership with DOE deploys Gemini across 17 US National Labs." }, { "id": "openai", "name": "OpenAI", "tier": "tier5", "category": "Frontier AI Lab", "website": "https://openai.com", "description": "Frontier model developer. April 2026 Industrial Policy paper explicitly calls for distributed AI-enabled laboratories as public infrastructure.", "funding": "$40B+ raised", "valuation": "$500B", "hq": "San Francisco, CA", "founded": "2015", "investors": "Microsoft, Thrive, Founders Fund", "differentiator": "Deep Research agent. Explicit policy focus on autonomous science, though no dedicated science product yet." }, { "id": "isomorphic-labs", "name": "Isomorphic Labs", "tier": "tier3", "category": "AI-Native Biotech", "website": "https://www.isomorphiclabs.com", "description": "Alphabet-backed AI drug discovery company spun out of Google DeepMind. Builds on AlphaFold to design novel therapeutics, with a 17-program pipeline advancing toward the clinic.", "funding": "$600M (first external round)", "valuation": "Undisclosed", "hq": "London, UK", "founded": "2021", "investors": "Thrive Capital, GV, Alphabet", "differentiator": "Founded and led by Demis Hassabis. Nearly $3B in collaborations with Eli Lilly and Novartis. IsoDDE design engine surpasses AlphaFold 3 at protein-ligand prediction; first AI-designed oncology drug targeting Phase 1 trials in 2026." }, { "id": "edison-scientific", "name": "Edison Scientific", "tier": "tier1", "category": "Autonomous Researcher", "website": "https://edisonscientific.com", "description": "Commercial spinout of FutureHouse. Kosmos is a multi-agent AI co-scientist that runs 12-hour cycles of literature review, data analysis, and hypothesis generation — executing ~42,000 lines of code and reading ~1,500 papers per run. Its 2026 surface also includes PaperQA3-backed Edison Literature and LABBench2.", "funding": "$70M seed", "valuation": "$250M", "hq": "San Francisco, CA", "founded": "2025", "investors": "Spark Capital, Triatomic Capital; angels include Jeff Dean and Dmitri Alperovitch", "differentiator": "CEO Sam Rodriques and co-founder Andrew White. Kosmos launch paper (arXiv 2511.02824) demonstrated seven discoveries across neuroscience, materials, and statistical genetics — three reproducing human-led findings, four entirely novel — while 2026 updates deepened the platform's literature and benchmark layer." }, { "id": "insilico-medicine", "name": "Insilico Medicine", "tier": "tier3", "category": "AI-Native Biotech", "website": "https://insilico.com", "description": "End-to-end generative AI platform for drug discovery and development (Pharma.AI). Public on the Hong Kong Stock Exchange since December 2025. TNIK inhibitor rentosertib is the first AI-discovered and AI-designed drug to deliver Phase 2a clinical data.", "funding": "Public (HKG: 03696.HK)", "valuation": "~$2B", "hq": "Hong Kong (also Boston, New York)", "founded": "2014", "investors": "Lilly, Tencent, Temasek, Schroders, UBS, Oaktree, E Fund, Taikang Life", "differentiator": "First AI-driven biotech to list on HKEX. Largest Hong Kong biotech IPO of 2025, oversubscribed 1,427× retail. $2.75B Eli Lilly partnership plus $888M Servier oncology deal. Founded by Alex Zhavoronkov." }, { "id": "ami-labs", "name": "AMI Labs", "tier": "tier5", "category": "Frontier AI Lab", "website": "https://amilabs.xyz", "description": "Advanced Machine Intelligence — Yann LeCun's new lab. Building world models based on Joint Embedding Predictive Architecture (JEPA) rather than next-token LLMs; aims to learn from physical reality as the path to human-level intelligence.", "funding": "$1.03B seed", "valuation": "$3.5B pre-money", "hq": "Paris, France", "founded": "2025", "investors": "Bezos Expeditions, Cathay Innovation, Greycroft, Hiro Capital, HV Capital, Nvidia, Temasek, Daphni, SBVA", "differentiator": "Largest seed round in European tech history. Turing Award winner Yann LeCun at the helm after leaving Meta; Saining Xie as CSO, Pascale Fung as chief research officer, Michael Rabbat as VP of world models. Explicit bet against the LLM paradigm." }, { "id": "chemlex", "name": "ChemLex", "tier": "tier2", "category": "Programmable Chemistry", "website": "https://chemlex.com", "description": "AI-for-science startup running a 24/7 autonomous chemistry synthesis line. Serves as the chemistry discovery engine for pharma; global HQ and flagship self-driving lab in Singapore.", "funding": "~$71M (Series A + Series B)", "valuation": "Undisclosed", "hq": "Singapore", "founded": "2022", "investors": "Granite Asia (lead), Qiming Venture Partners, LYFE Capital, Sinovation Ventures", "differentiator": "70+ customers including six of the top ten global pharma companies. MoU with Singapore's Experimental Drug Development Centre (EDDC). Incubated from MegaRobo; first major AI-for-science lab anchored in Asia." }, { "id": "coscientist-cmu", "name": "Coscientist (CMU)", "tier": "tier2", "category": "Autonomous Researcher", "website": "https://www.nature.com/articles/s41586-023-06792-0", "description": "Academic autonomous chemistry system from Carnegie Mellon. Uses GPT-4 with tool use, code execution, documentation search, and cloud-lab automation to design, plan, and execute chemistry experiments.", "funding": "Academic research", "hq": "Pittsburgh, PA", "founded": "2023", "investors": "Carnegie Mellon University; Emerald Cloud Lab collaboration", "differentiator": "Landmark Nature demonstration of an LLM-driven chemistry agent running real palladium cross-coupling experiments through Emerald Cloud Lab." }, { "id": "kebotix", "name": "Kebotix", "tier": "tier3", "category": "AI-Native Materials", "website": "https://www.kebotix.com", "description": "Self-driving laboratory company for chemicals and materials discovery. Combines AI planning, robotics, and high-throughput experimentation to shorten materials and molecule optimization loops.", "funding": "$11.4M Series A", "hq": "Cambridge, MA", "founded": "2017", "investors": "Novo Holdings, One Way Ventures, Baidu Ventures, Embark Ventures, Flybridge Capital", "differentiator": "Early commercial self-driving-lab pioneer focused on green chemistry and materials discovery rather than only informatics." }, { "id": "citrine-informatics", "name": "Citrine Informatics", "tier": "tier3", "category": "AI-Native Materials", "website": "https://citrine.io", "description": "Materials AI platform for chemicals, formulations, and manufactured products. Provides domain-specific data infrastructure and AI models that help R&D teams predict, optimize, and reuse materials knowledge.", "funding": "$43M+ disclosed", "hq": "Redwood City, CA", "founded": "2013", "investors": "Prelude Ventures, Innovation Endeavors, DCVC, Next47, Drive Catalyst", "differentiator": "One of the earliest enterprise materials-informatics platforms, with deployments at global chemicals and manufacturing companies." }, { "id": "artificial", "name": "Artificial", "tier": "tier3", "category": "Self-Driving Lab Software", "website": "https://www.artificial.com", "description": "Lab orchestration software for life sciences R&D. Connects wet-lab instruments, informatics systems, scheduling, digital twins, and AI agents into controlled scientific workflows.", "funding": "$21.5M Series A", "hq": "Palo Alto, CA", "founded": "2019", "investors": "Microsoft M12, Playground Global, AME Cloud Ventures", "differentiator": "Software-first control layer for robotic and human-operated labs, built to coordinate the full design-make-test-analyze cycle." }, { "id": "synthace", "name": "Synthace", "tier": "tier3", "category": "Self-Driving Lab Software", "website": "https://www.synthace.com", "description": "Cloud experiment-design platform for biological R&D. Turns multivariate experiment designs into automation instructions and structured data that can be analyzed and reused.", "funding": "$35M Series C", "hq": "London, UK", "founded": "2011", "investors": "Horizons Ventures, Sofinnova Partners, SOSV, Bioeconomy Capital", "differentiator": "No-code bridge between design-of-experiments planning, liquid-handling automation, and AI-ready experimental data." }, { "id": "tetrascience", "name": "TetraScience", "tier": "tier3", "category": "Self-Driving Lab Software", "website": "https://www.tetrascience.com", "description": "Scientific Data and AI Cloud for life sciences R&D. Replatforms raw instrument and experiment data into engineered, compliant datasets for analytics, automation, and scientific AI.", "funding": "$92M+ disclosed", "hq": "Boston, MA", "founded": "2014", "investors": "Insight Partners, Alkeon Capital, Waters Corporation", "differentiator": "Vendor-neutral data layer for AI-ready science, with broad lab-instrument and informatics integrations across enterprise pharma." }, { "id": "riffyn", "name": "Riffyn", "tier": "tier4", "category": "Self-Driving Lab Software", "website": "https://plm.sw.siemens.com/en-US/riffyn-x/", "description": "Scientific process design and data platform now offered as Riffyn X inside Siemens. Captures experimental methods, sample lineage, and process data in forms ready for analysis and machine learning.", "funding": "$19M+ disclosed before Siemens/Dotmatics integration", "hq": "Oakland, CA", "founded": "2014", "investors": "M Ventures, Waters, Siemens Venture Capital, OATV", "differentiator": "Process-centric experiment model that links recipe design, execution, chain of custody, and machine-learning-ready data." }, { "id": "opentrons", "name": "Opentrons", "tier": "tier3", "category": "Lab Robotics", "website": "https://opentrons.com", "description": "Open-source liquid-handling robotics company. Flex and OT-2 robots make programmable wet-lab automation accessible through modular hardware, Python APIs, and protocol libraries.", "funding": "Private, venture-backed", "hq": "New York, NY", "founded": "2013", "investors": "Khosla Ventures and others", "differentiator": "Low-cost, open robotics stack that lets smaller biology labs script and reproduce experiments without legacy automation overhead." }, { "id": "tecan", "name": "Tecan", "tier": "tier4", "category": "Lab Robotics", "website": "https://www.tecan.com", "description": "Swiss public lab-automation and life-science instrumentation company. Fluent and related workstations provide liquid handling, robotics, readers, and process automation across pharma and diagnostics.", "funding": "Public (SIX: TECN)", "hq": "Maennedorf, Switzerland", "founded": "1980", "investors": "Public market", "differentiator": "Industry-standard liquid-handling infrastructure that many AI-ready and self-driving lab stacks build around." }, { "id": "highres-biosolutions", "name": "HighRes Biosolutions", "tier": "tier3", "category": "Lab Robotics", "website": "https://www.highres.com", "description": "Life-science lab automation company combining modular hardware, liquid handling, dynamic scheduling, and Cellario orchestration software for drug discovery, genomics, and synthetic biology workflows.", "funding": "Private; Axel Johnson portfolio company", "hq": "Beverly, MA", "founded": "2004", "investors": "Axel Johnson", "differentiator": "Cellario and Nucleus turn heterogeneous instruments and workcells into schedulable, data-producing automation systems." }, { "id": "generate-biomedicines", "name": "Generate Biomedicines", "tier": "tier2", "category": "AI-Native Biotech", "website": "https://generatebiomedicines.com", "description": "Generative biology company creating programmable protein therapeutics. Combines machine-learning design with biological engineering and high-throughput wet-lab validation to build a clinical pipeline.", "funding": "Nearly $700M private financing before IPO path", "hq": "Somerville, MA", "founded": "2018", "investors": "Flagship Pioneering, Amgen, NVIDIA NVentures, ARCH, T. Rowe Price, ADIA", "differentiator": "Flagship platform company using learned protein design to generate antibodies, peptides, enzymes, and other therapeutic proteins." }, { "id": "insitro", "name": "Insitro", "tier": "tier2", "category": "AI-Native Biotech", "website": "https://www.insitro.com", "description": "Machine-learning drug discovery company founded by Daphne Koller. Builds large biological datasets from in vitro disease models, human genetics, and partnerships to find targets and patient subgroups.", "funding": "$600M+ disclosed", "valuation": "~$2.5B reported", "hq": "South San Francisco, CA", "founded": "2018", "investors": "Andreessen Horowitz, ARCH, GV, BlackRock, Temasek, CPP Investments", "differentiator": "Combines automated biology, human genetics, and ML under one roof rather than selling a narrow model or CRO service." }, { "id": "cellarity", "name": "Cellarity", "tier": "tier3", "category": "AI-Native Biotech", "website": "https://cellarity.com", "description": "Flagship Pioneering company developing cell-state-correcting medicines. Uses single-cell technologies, network biology, and machine learning to map disease cell behavior and design interventions.", "funding": "$274M disclosed", "hq": "Somerville, MA", "founded": "2017", "investors": "Flagship Pioneering, Kyowa Kirin, Hanwha Impact Partners", "differentiator": "Platform shifts discovery from single targets to whole-cell state transitions, with AI and multi-omics as the operating substrate." }, { "id": "abcellera", "name": "AbCellera", "tier": "tier3", "category": "Platform Biotech", "website": "https://www.abcellera.com", "description": "Public antibody-discovery platform company. Integrates single-cell screening, computation, antibody engineering, translational science, and clinical manufacturing from target to clinic.", "funding": "Public (NASDAQ: ABCL)", "hq": "Vancouver, Canada", "founded": "2012", "investors": "Public market; DARPA and pharma partnerships", "differentiator": "Full-stack antibody engine with in-house clinical manufacturing and a pandemic-response track record with Eli Lilly." }, { "id": "iambic-therapeutics", "name": "Iambic Therapeutics", "tier": "tier3", "category": "AI-Native Biotech", "website": "https://www.iambic.ai", "description": "AI-driven small-molecule drug discovery company. Uses multimodal transformers, protein-ligand structure prediction, high-throughput chemistry, and weekly biological data loops to advance candidates.", "funding": "$153M+ disclosed", "hq": "San Diego, CA", "founded": "2019", "investors": "Ascenta Capital, Abingworth, NVIDIA, Illumina Ventures, Coatue, OrbiMed, Sequoia", "differentiator": "Weekly design-make-test cycles combine AI-generated molecular designs with automated experimental execution." }, { "id": "causaly", "name": "Causaly", "tier": "tier3", "category": "Autonomous Researcher", "website": "https://www.causaly.com", "description": "Agentic AI platform for life-sciences R&D. Combines biomedical knowledge graphs, scientific retrieval, and governed agents for target identification, biomarker research, evidence review, and portfolio decisions.", "funding": "$93M+ disclosed", "hq": "London, UK; San Francisco, CA", "founded": "2018", "investors": "ICONIQ Growth, Index Ventures, Marathon Venture Capital, EBRD", "differentiator": "Science-grade agentic research layer used by pharma teams where factual grounding and traceable evidence matter more than generic chat." }, { "id": "iktos", "name": "Iktos", "tier": "tier3", "category": "Programmable Chemistry", "website": "https://iktos.ai", "description": "Generative AI and robotics platform for small-molecule drug discovery. Makya designs molecules, Spaya plans synthesis routes, and robotics execute synthesis, purification, analysis, and biological testing.", "funding": "EUR15.5M Series A plus EIC support", "hq": "Paris, France", "founded": "2016", "investors": "M Ventures, Debiopharm Innovation Fund, Omnes Capital, European Innovation Council", "differentiator": "Integrated design-make-test-analyze stack for medicinal chemistry rather than a design-only software platform." }, { "id": "genesis-molecular-ai", "name": "Genesis Molecular AI", "tier": "tier3", "category": "AI-Native Biotech", "website": "https://www.genesis.ml", "description": "AI-first small-molecule drug discovery company formerly known as Genesis Therapeutics. GEMS combines generative AI, molecular simulation, and predictive models to design molecules for difficult targets.", "funding": "$280M+ disclosed", "hq": "Burlingame, CA", "founded": "2019", "investors": "Andreessen Horowitz, Fidelity, BlackRock, NVIDIA NVentures, T. Rowe Price, Menlo Ventures", "differentiator": "Stanford spinout using foundation-model and physics methods to iterate from virtual designs to experimental validation." }, { "id": "argonne-national-laboratory", "name": "Argonne National Laboratory", "tier": "tier4", "category": "National Lab", "website": "https://www.anl.gov/autonomous-discovery", "description": "DOE national lab with an explicit Autonomous Discovery initiative. Uses AI, robotics, simulations, data platforms, and self-driving chemistry tools to automate experiments and scientific decision loops.", "funding": "Government", "hq": "Lemont, IL", "founded": "1946", "investors": "US Department of Energy", "differentiator": "Public autonomous-discovery program spanning robotic chemists, Polybot materials labs, AI-driven protein design, and autonomous data infrastructure." }, { "id": "oak-ridge-national-laboratory", "name": "Oak Ridge National Laboratory", "tier": "tier4", "category": "National Lab", "website": "https://www.ornl.gov/autonomousscience", "description": "DOE national lab building autonomous science platforms across instrumentation, computing, and laboratory automation. INTERSECT and the Autonomous Chemistry Lab connect AI, robotics, and HPC.", "funding": "Government", "hq": "Oak Ridge, TN", "founded": "1943", "investors": "US Department of Energy", "differentiator": "Combines leadership-class computing, user facilities, and self-driving labs under the Labs of the Future / Genesis Mission umbrella." }, { "id": "lawrence-livermore-national-laboratory", "name": "Lawrence Livermore National Laboratory", "tier": "tier4", "category": "National Lab", "website": "https://www.llnl.gov", "description": "DOE/NNSA national lab applying AI, robotics, and high-performance computing to scientific discovery. APEX is a self-driving alloy-discovery platform for autonomous design, fabrication, and characterization.", "funding": "Government", "hq": "Livermore, CA", "founded": "1952", "investors": "US Department of Energy / NNSA", "differentiator": "APEX targets the full materials loop for 3D-printed alloys, from model-guided design through robotic sample preparation and testing." }, { "id": "nist", "name": "NIST", "tier": "tier4", "category": "National Lab", "website": "https://www.nist.gov/autonomous-laboratories", "description": "US standards and measurement institute with autonomous-laboratory programs in materials, formulations, metrology, and biofoundry automation. Focuses on closed-loop experimentation and interoperability.", "funding": "Government", "hq": "Gaithersburg, MD", "founded": "1901", "investors": "US Department of Commerce", "differentiator": "Standards-oriented role for autonomous labs, including CAMEO, Autonomous Formulation Lab, Hermes software, and measurement-focused self-driving systems." }, { "id": "nrel", "name": "NREL", "tier": "tier4", "category": "National Lab", "website": "https://www.nrel.gov/materials-science/autonomous-experimentation", "description": "DOE national lab with autonomous experimentation capabilities for energy materials. Work includes autonomous synthesis, autonomous characterization, and AI-assisted control software for instruments.", "funding": "Government", "hq": "Golden, CO", "founded": "1977", "investors": "US Department of Energy", "differentiator": "Energy-materials focus: autonomous sputter deposition, autonomous microscopy, and open control-code workflows for real-world materials instruments." }, { "id": "brookhaven-national-laboratory", "name": "Brookhaven National Laboratory", "tier": "tier4", "category": "National Lab", "website": "https://www.bnl.gov", "description": "DOE national lab and user-facility operator applying AI, robotics, and autonomous experiment steering to materials and photon-science workflows at NSLS-II and CFN.", "funding": "Government", "hq": "Upton, NY", "founded": "1947", "investors": "US Department of Energy", "differentiator": "Large-scale user-facility angle: autonomous steering, Bluesky/robotics infrastructure, and AI-driven nanostructure discovery at Brookhaven facilities." }, { "id": "materials-genome-initiative", "name": "Materials Genome Initiative", "tier": "tier4", "category": "Academic Network", "website": "https://www.mgi.gov", "description": "US federal multi-agency initiative for accelerated materials discovery and deployment. Coordinates infrastructure, data, autonomous experimentation, and policy for materials innovation.", "funding": "Government program", "hq": "United States", "founded": "2011", "investors": "US federal agencies", "differentiator": "Foundational policy and coordination layer behind US materials-acceleration infrastructure, including autonomous experimentation roadmapping." }, { "id": "chai-discovery", "name": "Chai Discovery", "tier": "tier2", "category": "AI-Native Biotech", "website": "https://www.chaidiscovery.com", "description": "AI drug and biologics design company building a computer-aided design suite for molecules, with Chai-2 focused on zero-shot antibody and protein-binder design.", "funding": "$225M+", "valuation": "$1.3B", "hq": "San Francisco, CA", "founded": "2024", "investors": "Oak HC/FT, General Catalyst, Menlo Ventures, OpenAI, Thrive Capital, Dimension", "differentiator": "Chai-2 reports double-digit experimental success rates for de novo antibody design, a 100-fold improvement over previous computational methods, and a Lilly collaboration using proprietary Lilly data." }, { "id": "converge-bio", "name": "Converge Bio", "tier": "tier3", "category": "AI-Native Biotech", "website": "https://converge-bio.com", "description": "Generative AI platform for drug discovery and development, spanning target discovery, antibody design, and protein manufacturing optimization.", "funding": "$30M", "hq": "Boston, MA / Tel Aviv, Israel", "founded": "2024", "investors": "Bessemer Venture Partners, TLV Partners, Vintage Investment Partners, Saras Capital", "differentiator": "Over a dozen pharma and biotech customers and 40+ completed programs; models plug into existing workflows and are trained on large-scale proprietary and curated public datasets." }, { "id": "generare", "name": "Generare", "tier": "tier3", "category": "AI-Native Natural Products", "website": "https://generare.bio", "description": "Techbio company decoding microbial genomes to express silent chemistry and generate proprietary novel small-molecule data for drug discovery.", "funding": "€25M", "hq": "Paris, France", "founded": "2023", "investors": "Alven, Daphni, Galion.exe, Teampact Ventures, VIVES Partners", "differentiator": "Reported 200 previously uncharacterized small molecules in 2025 and plans to scale its proprietary molecular dataset ten-fold by 2027." }, { "id": "alloy-therapeutics", "name": "Alloy Therapeutics", "tier": "tier4", "category": "Biotech Infrastructure", "website": "https://www.alloytx.com", "description": "Tech-enabled biotech infrastructure and integrated services platform for biologics discovery and development, combining AI, real-world data, and wet-lab execution.", "funding": "$40M Series E", "valuation": "$1B", "hq": "Boston, MA", "founded": "2017", "investors": "8VC, JIC Venture Growth Investments, Mubadala Capital, Founders Fund, Alexandria Venture Investments", "differentiator": "200+ partners, 100+ licensed therapeutic programs, and 22 clinical programs; included as scope-review biotech infrastructure rather than an autonomous-lab operator." }, { "id": "robochem-flex", "name": "RoboChem-Flex", "tier": "tier4", "category": "Academic Self-Driving Lab", "website": "https://www.nature.com/articles/s44160-026-01053-0", "description": "Academic low-cost modular self-driving laboratory for automated chemical reaction optimization, with Python control software and Bayesian optimization.", "funding": "Academic research", "hq": "Netherlands", "founded": "2026", "investors": "University research program", "differentiator": "Affordable SDL infrastructure validated across six case studies with closed-loop and human-in-the-loop optimization modes." }, { "id": "scivity-labs", "name": "Scivity Labs", "tier": "tier3", "category": "Autonomous Science Platform", "website": "https://scivity.org", "description": "Autonomous science platform focused on computational research workflows that design, execute, and verify scientific work end-to-end, with public emphasis on validation outcomes and evidence surfaces.", "funding": "Pre-seed, undisclosed", "hq": "Yerevan, Armenia", "founded": "2025", "investors": "Undisclosed", "differentiator": "Verification-first positioning: public outputs and evidence surfaces are open, while thresholds, calibration, and check-composition logic remain private." }, { "id": "tetsuwan-scientific", "name": "Tetsuwan Scientific", "tier": "tier2", "category": "Self-Driving Lab Software", "website": "https://tetsuwan.com", "description": "Execution-layer startup building ResearchOS and a future cloud lab so researchers or AI agents can translate natural-language procedures into machine-ready wet-lab workflows.", "funding": "$2.7M pre-seed", "hq": "San Francisco, CA", "founded": "2023", "investors": "2048 Ventures, Carbon Silicon Ventures, Referent Ventures, Everywhere Ventures", "differentiator": "ResearchOS positions itself as the interface between researchers, agents, and automated biology labs, with cloud-lab ambitions from day one." }, { "id": "microsoft-discovery", "name": "Microsoft Discovery", "tier": "tier5", "category": "Enterprise R&D Platform", "website": "https://azure.microsoft.com/en-us/solutions/discovery", "description": "Microsoft's enterprise agentic R&D platform for hypothesis generation, simulation, analysis, and iterative experimentation across life sciences, materials, semiconductors, and engineering workflows.", "funding": "Microsoft-funded preview", "hq": "Redmond, WA", "founded": "2025", "investors": "Microsoft", "differentiator": "Pairs a graph-based Discovery Engine with Azure governance, HPC, and partner integrations to move from scientific reasoning to governed enterprise execution." }, { "id": "anthropic", "name": "Anthropic", "tier": "tier5", "category": "Frontier AI Lab", "website": "https://www.anthropic.com", "description": "Frontier model developer (Claude family). October 2025 launched Claude for Life Sciences with the Allen Institute and HHMI as founding scientific partners and connectors to Benchling, PubMed, 10x Genomics, ClinicalTrials.gov, and Synapse. April 2026 acquired Coefficient Bio (all-stock, ~$400M) as its first M&A into a new domain. Top of the Ai2 AstaBench scientific-agent leaderboard (Claude Opus 4.7 at 58.0%).", "funding": "$10B+ raised across rounds", "valuation": "$183B (reported 2025)", "hq": "San Francisco, CA", "founded": "2021", "investors": "Amazon, Google, Lightspeed, Spark Capital, Salesforce, Menlo Ventures", "differentiator": "Best-in-class scientific reasoning on AstaBench paired with the first dedicated frontier-lab life-sciences product and a Coefficient Bio acqui-hire to anchor the science team." }, { "id": "ai2", "name": "Ai2", "tier": "tier2", "category": "Autonomous Researcher", "website": "https://allenai.org", "description": "Allen Institute for AI — nonprofit research institute founded by Paul Allen. In August 2025 launched Asta, an open agentic ecosystem for science covering scholarly research assistants, the AstaBench benchmark suite (2,400+ problems across 11 benchmarks), and a developer toolkit with open-source agents, post-trained science LMs, and a 200M-paper Scientific Corpus Tool. AstaBench accepted as ICLR 2026 oral.", "funding": "Endowment-backed nonprofit", "hq": "Seattle, WA", "founded": "2014", "investors": "Paul G. Allen Family Foundation (founding)", "differentiator": "Open scientific-agent stack and the de-facto benchmark for scientific AI agents, already adopted by UK AISI, General Reasoning, Elicit, SciSpace, Distyl AI, and EvoScientist." }, { "id": "amazon-bio-discovery", "name": "Amazon Bio Discovery", "tier": "tier5", "category": "Enterprise R&D Platform", "website": "https://aws.amazon.com/health/biology-foundation-models/", "description": "AWS's agentic AI application for drug research, announced April 28, 2026. Exposes 40+ biology foundation models, supports custom uploads and fine-tuning, and integrates with CRO and cloud-lab partners (Ginkgo Nebula, Twist Bioscience, with A-Alpha Bio anticipated) so that agents select BioFMs, route candidates to wet-lab partners for synthesis and testing, and return experimental results into the loop on Bedrock.", "funding": "Amazon-funded service", "hq": "Seattle, WA", "founded": "2026", "investors": "Amazon", "differentiator": "First Big Tech lab-in-the-loop platform that closes the agent → BioFM → wet-lab → result cycle directly on AWS infrastructure, with an early Memorial Sloan Kettering project compressing a year of antibody triage into weeks." }, { "id": "profluent", "name": "Profluent", "tier": "tier2", "category": "AI-Native Biotech", "website": "https://www.profluent.bio", "description": "Programmable-biology company using frontier protein language models. November 2025 closed a $106M Series B co-led by Altimeter Capital and Bezos Expeditions, taking total funding to $150M. May 2026 announced a strategic collaboration with Eli Lilly for custom site-specific recombinases addressing severe-unmet-need diseases, with up to $2.25B in development and commercial milestones plus tiered royalties. Released OpenCRISPR-1, the first AI-generated CRISPR gene editor.", "funding": "$150M total ($106M Series B Nov 2025)", "hq": "Berkeley, CA", "founded": "2022", "investors": "Altimeter Capital, Bezos Expeditions, Spark Capital, Insight Partners, Air Street Capital", "differentiator": "Demonstrated scaling laws for protein design (NeurIPS 2025 spotlight) and the first wholly AI-designed CRISPR system (Nature 2025), then converted that into a $2.25B Lilly recombinase pact in May 2026." } ], "entityDetails": { "Periodic Labs": { "id": "periodic-labs", "entityType": "Company", "stage": "Private scaleup", "region": "North America", "scienceDomains": [ "Materials", "Chemistry" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Autonomous materials labs", "AI scientist workflow", "Experimental data flywheel" ], "scores": { "autonomy": 4, "execution": 4, "verification": 3, "scientificEvidence": 3, "capital": 4, "traction": 2 }, "evidenceLevel": "Announced platform", "accessModel": "Proprietary", "tractionSignals": [ "$300M seed", "Frontier-lab founding team", "Materials-first autonomous lab buildout" ], "sourceIds": [ "official-periodic-labs" ] }, "Lila Sciences": { "id": "lila-sciences", "entityType": "Company", "stage": "Private scaleup", "region": "North America", "scienceDomains": [ "Biology", "Chemistry", "Materials" ], "stackRoles": [ "Reason", "Execute", "Verify", "Data" ], "products": [ "AI Science Factories", "Reasoning models", "Robotic lab network" ], "scores": { "autonomy": 4, "execution": 4, "verification": 3, "scientificEvidence": 3, "capital": 4, "traction": 2 }, "evidenceLevel": "Announced platform", "accessModel": "Proprietary", "tractionSignals": [ "$550M raised", "Flagship Pioneering origin", "Life, chemical, and materials scope" ], "sourceIds": [ "official-lila-sciences" ] }, "Sakana AI": { "id": "sakana-ai", "entityType": "Company", "stage": "Private scaleup", "region": "Asia-Pacific", "scienceDomains": [ "ML research", "General science" ], "stackRoles": [ "Reason", "Verify" ], "products": [ "AI Scientist", "AI Scientist-v2", "Marlin" ], "scores": { "autonomy": 4, "execution": 1, "verification": 2, "scientificEvidence": 3, "capital": 4, "traction": 3 }, "evidenceLevel": "Peer-reviewed demo", "accessModel": "Hybrid", "tractionSignals": [ "AI-authored paper workflow", "$379M cumulative funding", "Commercial enterprise research beta" ], "sourceIds": [ "official-sakana-ai" ] }, "FutureHouse": { "id": "futurehouse", "entityType": "Nonprofit lab", "stage": "Research platform", "region": "North America", "scienceDomains": [ "Biology", "Literature", "General science" ], "stackRoles": [ "Reason", "Execute", "Verify", "Data" ], "products": [ "Crow", "Falcon", "Owl", "Phoenix", "PaperQA2", "Robin", "ether0", "DISCO" ], "scores": { "autonomy": 4, "execution": 1, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Open research", "accessModel": "Hybrid", "tractionSignals": [ "Open scientific agents", "LitQA2 performance", "DISCO wet-lab validation", "Commercial spinout via Edison Scientific" ], "sourceIds": [ "official-futurehouse", "official-futurehouse-disco-202604" ] }, "Autoscience Institute": { "id": "autoscience-institute", "entityType": "Company", "stage": "Seed", "region": "North America", "scienceDomains": [ "ML research" ], "stackRoles": [ "Reason", "Verify" ], "products": [ "Carl", "Mira" ], "scores": { "autonomy": 4, "execution": 1, "verification": 2, "scientificEvidence": 3, "capital": 2, "traction": 2 }, "evidenceLevel": "Peer-reviewed demo", "accessModel": "Proprietary", "tractionSignals": [ "$14M seed", "ICLR workshop acceptances", "Live Kaggle competition entries" ], "sourceIds": [ "official-autoscience-institute" ] }, "Thesis": { "id": "thesis", "entityType": "Company", "stage": "Seed", "region": "North America", "scienceDomains": [ "ML research" ], "stackRoles": [ "Reason", "Verify" ], "products": [ "Automated ML research workflow" ], "scores": { "autonomy": 3, "execution": 1, "verification": 2, "scientificEvidence": 2, "capital": 1, "traction": 1 }, "evidenceLevel": "Benchmark result", "accessModel": "Proprietary", "tractionSignals": [ "YC F25", "MLE-Bench medal-rate claim", "Stanford AI Lab founding network" ], "sourceIds": [ "official-thesis" ] }, "Zeon Systems": { "id": "zeon-systems", "entityType": "Company", "stage": "Seed", "region": "North America", "scienceDomains": [ "Lab automation", "Biology" ], "stackRoles": [ "Execute", "Infrastructure" ], "products": [ "Natural-language lab robotics layer", "Labware vision system" ], "scores": { "autonomy": 2, "execution": 3, "verification": 1, "scientificEvidence": 1, "capital": 1, "traction": 1 }, "evidenceLevel": "Pilot", "accessModel": "Proprietary", "tractionSignals": [ "YC X25", "UCSF and Stanford pilots", "Robot-arm protocol execution" ], "sourceIds": [ "official-zeon-systems" ] }, "Automata": { "id": "automata", "entityType": "Company", "stage": "Growth", "region": "Europe", "scienceDomains": [ "Biology", "Diagnostics", "Lab automation" ], "stackRoles": [ "Execute", "Infrastructure", "Data" ], "products": [ "LINQ", "Automated lab benches", "LINQ Cloud" ], "scores": { "autonomy": 2, "execution": 4, "verification": 2, "scientificEvidence": 2, "capital": 2, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "Five top pharma customers", "$45M Series C", "Danaher strategic participation" ], "sourceIds": [ "official-automata" ] }, "Cenevo": { "id": "cenevo", "entityType": "Company", "stage": "Commercial", "region": "North America / Europe", "scienceDomains": [ "Biology", "Lab automation", "Scientific informatics" ], "stackRoles": [ "Execute", "Data", "Infrastructure" ], "products": [ "Labguru ELN/LIMS", "Titian Mosaic sample management", "AI Protocol Conversion", "AI Automation agent" ], "scores": { "autonomy": 2, "execution": 3, "verification": 2, "scientificEvidence": 2, "capital": 2, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "950+ customer organizations", "45,000+ users", "8 of the top 10 pharma companies" ], "sourceIds": [ "official-cenevo-rebrand-202507", "rdworld-cenevo-ai-agents-202602" ] }, "Emerald Cloud Lab": { "id": "emerald-cloud-lab", "entityType": "Company", "stage": "Commercial", "region": "North America", "scienceDomains": [ "Chemistry", "Biology", "Cloud lab" ], "stackRoles": [ "Execute", "Infrastructure", "Data", "Verify" ], "products": [ "Remote programmable lab", "Scientific computing environment" ], "scores": { "autonomy": 2, "execution": 4, "verification": 3, "scientificEvidence": 3, "capital": 2, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "Mature programmable cloud lab", "Academic and pharma users", "Remote experiment execution" ], "sourceIds": [ "official-emerald-cloud-lab" ] }, "Strateos": { "id": "strateos", "entityType": "Company", "stage": "Commercial", "region": "North America", "scienceDomains": [ "Biology", "Chemistry", "Cloud lab" ], "stackRoles": [ "Execute", "Infrastructure", "Data" ], "products": [ "Robotic cloud lab", "API experiment orchestration", "On-prem deployments" ], "scores": { "autonomy": 2, "execution": 4, "verification": 2, "scientificEvidence": 2, "capital": 2, "traction": 3 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "Pharma R&D focus", "$56M funding", "API-first orchestration" ], "sourceIds": [ "official-strateos" ] }, "Chemify": { "id": "chemify", "entityType": "Company", "stage": "Growth", "region": "Europe", "scienceDomains": [ "Chemistry", "Drug discovery", "Programmable synthesis" ], "stackRoles": [ "Reason", "Execute", "Verify", "Infrastructure" ], "products": [ "Chemputer", "Chemifarm", "Reaction planning software" ], "scores": { "autonomy": 3, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 3, "traction": 3 }, "evidenceLevel": "Published platform", "accessModel": "Proprietary", "tractionSignals": [ "Chemistry-as-code pedigree", "Series B expansion", "Peer-reviewed validation" ], "sourceIds": [ "official-chemify" ] }, "Recursion Pharmaceuticals": { "id": "recursion-pharmaceuticals", "entityType": "Public company", "stage": "Public", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery", "Imaging" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Recursion OS", "Industrial phenomics platform", "BioHive supercomputer" ], "scores": { "autonomy": 2, "execution": 3, "verification": 3, "scientificEvidence": 4, "capital": 4, "traction": 4 }, "evidenceLevel": "Public company", "accessModel": "Proprietary", "tractionSignals": [ "NASDAQ-listed", "Exscientia merger", "Industrial-scale image-based screening" ], "sourceIds": [ "official-recursion-pharmaceuticals" ] }, "Atinary": { "id": "atinary", "entityType": "Company", "stage": "Commercial", "region": "Europe", "scienceDomains": [ "Chemistry", "Materials", "Drug discovery" ], "stackRoles": [ "Reason", "Execute", "Verify", "Data" ], "products": [ "Self-driving lab software", "Boston Scientific Discovery Factories", "Bayesian optimization workflows" ], "scores": { "autonomy": 3, "execution": 3, "verification": 3, "scientificEvidence": 3, "capital": 1, "traction": 2 }, "evidenceLevel": "Commercial hardware and software", "accessModel": "Proprietary", "tractionSignals": [ "Software-first self-driving lab approach", "Boston Scientific Discovery Factories announced in 2026", "Chemistry informatics partnerships", "Enterprise R&D focus" ], "sourceIds": [ "official-atinary", "official-atinary-boston-sdl-202602", "selectscience-atinary-boston-sdl-202602" ] }, "Ginkgo Bioworks": { "id": "ginkgo-bioworks", "entityType": "Public company", "stage": "Public", "region": "North America", "scienceDomains": [ "Biology", "Synthetic biology", "Cloud lab" ], "stackRoles": [ "Reason", "Execute", "Data", "Infrastructure" ], "products": [ "Foundry", "Nebula", "Ginkgo Cloud Lab", "EstiMate" ], "scores": { "autonomy": 3, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 3, "traction": 4 }, "evidenceLevel": "Public company", "accessModel": "Hybrid", "tractionSignals": [ "NYSE-listed", "Nebula autonomous lab", "Large biological data generation platform" ], "sourceIds": [ "official-ginkgo-bioworks" ] }, "Radical AI": { "id": "radical-ai", "entityType": "Company", "stage": "Seed", "region": "North America", "scienceDomains": [ "Materials", "Simulation" ], "stackRoles": [ "Reason", "Execute", "Data" ], "products": [ "Materials foundation models", "TorchSim", "Autonomous materials lab" ], "scores": { "autonomy": 3, "execution": 3, "verification": 2, "scientificEvidence": 2, "capital": 2, "traction": 2 }, "evidenceLevel": "Open research", "accessModel": "Hybrid", "tractionSignals": [ "TorchSim open-source release", "AFWERX contract", "Brooklyn autonomous materials lab" ], "sourceIds": [ "official-radical-ai" ] }, "Xaira Therapeutics": { "id": "xaira-therapeutics", "entityType": "Company", "stage": "Private scaleup", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery", "Virtual cell" ], "stackRoles": [ "Reason", "Data", "Verify" ], "products": [ "X-Atlas", "X-Cell", "AI drug discovery pipeline" ], "scores": { "autonomy": 2, "execution": 2, "verification": 3, "scientificEvidence": 3, "capital": 4, "traction": 2 }, "evidenceLevel": "Published data asset", "accessModel": "Proprietary", "tractionSignals": [ "$1B+ Series A", "25.6M-cell atlas", "Nobel laureate founder network" ], "sourceIds": [ "official-xaira-therapeutics" ] }, "Helical": { "id": "helical", "entityType": "Company", "stage": "Seed", "region": "Europe", "scienceDomains": [ "Biology", "Drug discovery", "Biomarkers" ], "stackRoles": [ "Reason", "Data", "Verify" ], "products": [ "Virtual Lab", "Model Factory", "Bio foundation model workflows" ], "scores": { "autonomy": 3, "execution": 1, "verification": 3, "scientificEvidence": 2, "capital": 2, "traction": 3 }, "evidenceLevel": "Production deployment", "accessModel": "Proprietary", "tractionSignals": [ "$10M seed", "Production deployments with multiple top-20 pharma companies", "Public Pfizer collaboration on predictive blood-based safety biomarkers" ], "sourceIds": [ "official-helical", "globenewswire-helical-seed-202604", "techeu-helical-seed-202604" ] }, "Fathom Therapeutics": { "id": "fathom-therapeutics", "entityType": "Company", "stage": "Series A", "region": "North America", "scienceDomains": [ "Drug discovery", "Chemistry", "Biology" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Microcosmos", "AI-enabled small-molecule discovery engine", "Lab-in-the-loop discovery programs" ], "scores": { "autonomy": 2, "execution": 2, "verification": 2, "scientificEvidence": 3, "capital": 3, "traction": 3 }, "evidenceLevel": "Working prototype", "accessModel": "Proprietary", "tractionSignals": [ "$47M Series A", "Lead program in animal efficacy studies", "Multiple external discovery partnerships" ], "sourceIds": [ "official-fathom-therapeutics", "prnewswire-fathom-series-a-202604", "cen-fathom-series-a-202604" ] }, "10x Science": { "id": "10x-science", "entityType": "Company", "stage": "Seed", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery", "Protein characterization" ], "stackRoles": [ "Data", "Verify" ], "products": [ "AI-native protein characterization platform", "Mass spectrometry analysis agents", "Proteoform analysis workflows" ], "scores": { "autonomy": 2, "execution": 1, "verification": 3, "scientificEvidence": 2, "capital": 1, "traction": 2 }, "evidenceLevel": "Announced platform", "accessModel": "Proprietary", "tractionSignals": [ "$4.8M seed", "YC W26", "Stanford/Bertozzi-lab founder expertise" ], "sourceIds": [ "official-10x-science", "yc-10x-science-profile-202604", "techcrunch-10x-science-seed-202604", "prnewswire-10x-science-seed-202604" ] }, "A-Lab (LBNL)": { "id": "a-lab-lbnl", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Materials", "Chemistry" ], "stackRoles": [ "Reason", "Execute", "Verify", "Data" ], "products": [ "Autonomous materials synthesis facility" ], "scores": { "autonomy": 4, "execution": 4, "verification": 2, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Published platform", "accessModel": "Public infrastructure", "tractionSignals": [ "Nature-published autonomous synthesis", "DOE national lab context", "Defining verification case study" ], "sourceIds": [ "official-a-lab-lbnl" ] }, "Acceleration Consortium": { "id": "acceleration-consortium", "entityType": "Academic network", "stage": "Research network", "region": "North America", "scienceDomains": [ "Materials", "Chemistry", "Standards" ], "stackRoles": [ "Execute", "Verify", "Infrastructure", "Data" ], "products": [ "Self-driving lab network", "Protocols", "Standards coordination" ], "scores": { "autonomy": 2, "execution": 3, "verification": 3, "scientificEvidence": 4, "capital": 3, "traction": 3 }, "evidenceLevel": "Research network", "accessModel": "Public infrastructure", "tractionSignals": [ "$200M CFREF grant", "30+ member institutions", "Standards role" ], "sourceIds": [ "official-acceleration-consortium" ] }, "PNNL": { "id": "pnnl", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Chemistry", "Energy", "Materials" ], "stackRoles": [ "Execute", "Verify", "Infrastructure", "Data" ], "products": [ "Autonomy Studio", "DOE autonomous experimentation programs" ], "scores": { "autonomy": 2, "execution": 3, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Research facility", "accessModel": "Public infrastructure", "tractionSignals": [ "DOE national lab", "Autonomy Studio", "Genesis Mission network" ], "sourceIds": [ "official-pnnl" ] }, "Google DeepMind": { "id": "google-deepmind", "entityType": "Frontier AI lab", "stage": "Big Tech lab", "region": "Europe", "scienceDomains": [ "Biology", "Materials", "Mathematics", "General science" ], "stackRoles": [ "Reason", "Verify", "Data" ], "products": [ "AlphaFold", "AlphaProteo", "GNoME", "AlphaGeometry", "AI co-scientist" ], "scores": { "autonomy": 3, "execution": 1, "verification": 3, "scientificEvidence": 4, "capital": 4, "traction": 4 }, "evidenceLevel": "Frontier research", "accessModel": "Hybrid", "tractionSignals": [ "Nobel-recognized AlphaFold work", "DOE Genesis Mission partnership", "Deep science AI bench" ], "sourceIds": [ "official-google-deepmind", "epoch-capabilities" ] }, "OpenAI": { "id": "openai", "entityType": "Frontier AI lab", "stage": "Frontier platform", "region": "North America", "scienceDomains": [ "General science", "Biology", "Reasoning" ], "stackRoles": [ "Reason", "Verify", "Infrastructure" ], "products": [ "Deep Research", "o-series reasoning models", "GPT-Rosalind", "Prism" ], "scores": { "autonomy": 3, "execution": 1, "verification": 3, "scientificEvidence": 3, "capital": 4, "traction": 4 }, "evidenceLevel": "Frontier platform", "accessModel": "Proprietary", "tractionSignals": [ "DOE national labs access", "Science policy agenda", "Specialized biology model launch" ], "sourceIds": [ "official-openai", "artificial-analysis-leaderboard", "artificial-analysis-methodology" ] }, "Isomorphic Labs": { "id": "isomorphic-labs", "entityType": "Company", "stage": "Private scaleup", "region": "Europe", "scienceDomains": [ "Biology", "Drug discovery" ], "stackRoles": [ "Reason", "Verify", "Data" ], "products": [ "IsoDDE", "AI drug design platform" ], "scores": { "autonomy": 2, "execution": 2, "verification": 3, "scientificEvidence": 4, "capital": 4, "traction": 3 }, "evidenceLevel": "Clinical pipeline", "accessModel": "Proprietary", "tractionSignals": [ "$600M first external round", "Nearly $3B in pharma collaborations", "17-program pipeline" ], "sourceIds": [ "official-isomorphic-labs" ] }, "Edison Scientific": { "id": "edison-scientific", "entityType": "Company", "stage": "Seed", "region": "North America", "scienceDomains": [ "Biology", "Materials", "Statistical genetics", "General science" ], "stackRoles": [ "Reason", "Verify", "Data" ], "products": [ "Kosmos", "Edison Literature", "PaperQA3", "LABBench2" ], "scores": { "autonomy": 4, "execution": 1, "verification": 3, "scientificEvidence": 3, "capital": 3, "traction": 2 }, "evidenceLevel": "Launch paper", "accessModel": "Proprietary", "tractionSignals": [ "$70M seed", "Seven demonstrated discoveries", "PaperQA3 literature upgrade", "NVIDIA scaling partnership", "FutureHouse commercial spinout" ], "sourceIds": [ "official-edison-scientific", "official-edison-paperqa3-202602", "official-edison-nvidia-202603", "official-edison-labbench2-202605" ] }, "Insilico Medicine": { "id": "insilico-medicine", "entityType": "Public company", "stage": "Public", "region": "Asia-Pacific", "scienceDomains": [ "Biology", "Drug discovery", "Clinical development" ], "stackRoles": [ "Reason", "Verify", "Data" ], "products": [ "Pharma.AI", "rentosertib", "AI drug discovery pipeline" ], "scores": { "autonomy": 2, "execution": 2, "verification": 3, "scientificEvidence": 4, "capital": 4, "traction": 4 }, "evidenceLevel": "Clinical data", "accessModel": "Proprietary", "tractionSignals": [ "HKEX listing", "Phase 2a clinical data", "$2.75B Lilly partnership" ], "sourceIds": [ "official-insilico-medicine" ] }, "AMI Labs": { "id": "ami-labs", "entityType": "Frontier AI lab", "stage": "Seed", "region": "Europe", "scienceDomains": [ "World models", "General intelligence", "Physical reasoning" ], "stackRoles": [ "Reason", "Data" ], "products": [ "JEPA-first world model research agenda" ], "scores": { "autonomy": 2, "execution": 1, "verification": 1, "scientificEvidence": 1, "capital": 4, "traction": 1 }, "evidenceLevel": "Announced lab", "accessModel": "Proprietary", "tractionSignals": [ "$1.03B seed", "Yann LeCun founding team", "World-model-first thesis" ], "sourceIds": [ "official-ami-labs" ] }, "ChemLex": { "id": "chemlex", "entityType": "Company", "stage": "Growth", "region": "Asia-Pacific", "scienceDomains": [ "Chemistry", "Drug discovery", "Programmable synthesis" ], "stackRoles": [ "Reason", "Execute", "Infrastructure", "Data" ], "products": [ "24/7 autonomous synthesis line", "Chemistry CRO workflow" ], "scores": { "autonomy": 3, "execution": 4, "verification": 2, "scientificEvidence": 2, "capital": 3, "traction": 3 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "70+ customers", "Six top-ten pharma customers", "Singapore flagship self-driving lab" ], "sourceIds": [ "official-chemlex" ] }, "Coscientist (CMU)": { "id": "coscientist-cmu", "entityType": "Academic project", "stage": "Published research", "region": "North America", "scienceDomains": [ "Chemistry", "Autonomous agents", "Cloud lab" ], "stackRoles": [ "Reason", "Execute", "Verify", "Data" ], "products": [ "Coscientist autonomous chemistry system", "Cloud-lab experiment execution" ], "scores": { "autonomy": 4, "execution": 3, "verification": 3, "scientificEvidence": 4, "capital": 1, "traction": 2 }, "evidenceLevel": "Nature paper", "accessModel": "Research prototype", "tractionSignals": [ "Nature 2023 publication", "Emerald Cloud Lab execution", "Palladium cross-coupling demonstrations" ], "sourceIds": [ "cmu-coscientist-202312", "nature-coscientist-202312" ] }, "Kebotix": { "id": "kebotix", "entityType": "Company", "stage": "Commercial", "region": "North America", "scienceDomains": [ "Materials", "Chemistry", "Green chemistry" ], "stackRoles": [ "Reason", "Execute", "Data" ], "products": [ "Self-driving materials discovery platform", "AI-guided chemistry workflows" ], "scores": { "autonomy": 3, "execution": 3, "verification": 2, "scientificEvidence": 2, "capital": 2, "traction": 2 }, "evidenceLevel": "Commercial platform", "accessModel": "Proprietary", "tractionSignals": [ "$11.4M Series A", "Novo Holdings lead", "Self-driving-lab positioning" ], "sourceIds": [ "official-kebotix", "finsmes-kebotix-series-a-202004" ] }, "Citrine Informatics": { "id": "citrine-informatics", "entityType": "Company", "stage": "Growth", "region": "North America", "scienceDomains": [ "Materials", "Chemistry", "Manufacturing" ], "stackRoles": [ "Reason", "Data", "Verify" ], "products": [ "Citrine Platform", "Materials AI models", "R&D data infrastructure" ], "scores": { "autonomy": 2, "execution": 1, "verification": 3, "scientificEvidence": 3, "capital": 3, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "$16M Series C", "Global chemicals customers", "Materials AI operating-platform positioning" ], "sourceIds": [ "official-citrine-informatics", "businesswire-citrine-series-c-202301" ] }, "Artificial": { "id": "artificial", "entityType": "Company", "stage": "Series A", "region": "North America", "scienceDomains": [ "Lab automation", "Biology", "Synthetic biology" ], "stackRoles": [ "Execute", "Infrastructure", "Data" ], "products": [ "aLab Suite", "Lab orchestration platform", "Digital twin builder" ], "scores": { "autonomy": 3, "execution": 3, "verification": 2, "scientificEvidence": 2, "capital": 2, "traction": 2 }, "evidenceLevel": "Commercial platform", "accessModel": "Proprietary", "tractionSignals": [ "$21.5M Series A", "Thermo Fisher and Beam Therapeutics customers", "AI agents for lab process optimization" ], "sourceIds": [ "official-artificial", "techcrunch-artificial-series-a-202105" ] }, "Synthace": { "id": "synthace", "entityType": "Company", "stage": "Growth", "region": "Europe", "scienceDomains": [ "Biology", "Experiment design", "Lab automation" ], "stackRoles": [ "Reason", "Execute", "Data" ], "products": [ "Synthace platform", "Automated multivariate experiments", "Automation instruction generation" ], "scores": { "autonomy": 2, "execution": 3, "verification": 2, "scientificEvidence": 2, "capital": 3, "traction": 3 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "$35M Series C", "Top pharma and CRO customers", "No-code DOE-to-automation workflow" ], "sourceIds": [ "official-synthace", "finsmes-synthace-series-c-202111" ] }, "TetraScience": { "id": "tetrascience", "entityType": "Company", "stage": "Growth", "region": "North America", "scienceDomains": [ "Scientific data", "Biopharma", "Lab automation" ], "stackRoles": [ "Data", "Infrastructure", "Verify" ], "products": [ "Tetra Scientific Data and AI Platform", "Tetra Data Platform", "Tetra AI" ], "scores": { "autonomy": 2, "execution": 1, "verification": 3, "scientificEvidence": 2, "capital": 3, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "$80M Series B", "Enterprise pharma adoption", "Instrument and informatics partner network" ], "sourceIds": [ "official-tetrascience", "fierce-tetrascience-series-b-202104" ] }, "Riffyn": { "id": "riffyn", "entityType": "Company / product line", "stage": "Commercial", "region": "North America", "scienceDomains": [ "Bioprocessing", "Scientific data", "Experiment design" ], "stackRoles": [ "Data", "Verify", "Infrastructure" ], "products": [ "Riffyn X", "Scientific Development Environment", "Process data models" ], "scores": { "autonomy": 1, "execution": 1, "verification": 3, "scientificEvidence": 2, "capital": 2, "traction": 3 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "$15M Series B", "BASF collaboration", "Siemens / Dotmatics current product line" ], "sourceIds": [ "siemens-riffyn-x", "biospace-riffyn-series-b-201905" ] }, "Opentrons": { "id": "opentrons", "entityType": "Company", "stage": "Commercial", "region": "North America", "scienceDomains": [ "Lab automation", "Biology", "Liquid handling" ], "stackRoles": [ "Execute", "Infrastructure" ], "products": [ "Opentrons Flex", "OT-2", "Python Protocol API", "Protocol Designer" ], "scores": { "autonomy": 2, "execution": 4, "verification": 2, "scientificEvidence": 3, "capital": 2, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Hybrid", "tractionSignals": [ "Open-source software and firmware", "Flex launch", "Mayo Clinic, Harvard, and MIT customer signals" ], "sourceIds": [ "official-opentrons-flex", "techcrunch-opentrons-flex-202305" ] }, "Tecan": { "id": "tecan", "entityType": "Public company", "stage": "Public", "region": "Europe", "scienceDomains": [ "Lab automation", "Diagnostics", "Life sciences" ], "stackRoles": [ "Execute", "Infrastructure" ], "products": [ "Fluent", "Freedom EVO", "Tecan Labwerx", "Liquid handling systems" ], "scores": { "autonomy": 1, "execution": 4, "verification": 2, "scientificEvidence": 3, "capital": 3, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Commercial hardware", "tractionSignals": [ "SIX-listed", "3,000+ employees", "Global lab automation installed base" ], "sourceIds": [ "official-tecan-history", "stockanalysis-tecan-profile-202603" ] }, "HighRes Biosolutions": { "id": "highres-biosolutions", "entityType": "Company", "stage": "Commercial", "region": "North America", "scienceDomains": [ "Lab automation", "Drug discovery", "Synthetic biology", "Genomics" ], "stackRoles": [ "Execute", "Infrastructure", "Data" ], "products": [ "Cellario", "Nucleus", "Prime liquid handler", "AutoPod" ], "scores": { "autonomy": 2, "execution": 4, "verification": 2, "scientificEvidence": 2, "capital": 2, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Commercial hardware and software", "tractionSignals": [ "20+ years deployments", "Axel Johnson ownership", "Cellario OS orchestration platform" ], "sourceIds": [ "official-highres", "biodev-highres-cellario-202509" ] }, "Generate Biomedicines": { "id": "generate-biomedicines", "entityType": "Company", "stage": "Clinical-stage platform", "region": "North America", "scienceDomains": [ "Biology", "Protein design", "Drug discovery" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Generate Platform", "Generative Biology", "Protein therapeutic pipeline" ], "scores": { "autonomy": 2, "execution": 3, "verification": 3, "scientificEvidence": 3, "capital": 4, "traction": 3 }, "evidenceLevel": "Clinical pipeline", "accessModel": "Proprietary", "tractionSignals": [ "$370M Series B", "$273M Series C", "Amgen collaboration", "17-program pipeline reported in 2023" ], "sourceIds": [ "official-generate-biomedicines", "fierce-generate-series-b-202111", "businesswire-generate-series-c-202309" ] }, "Insitro": { "id": "insitro", "entityType": "Company", "stage": "Private scaleup", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery", "Human genetics" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Machine-learning drug discovery platform", "In vitro disease models", "Human genetics data platform" ], "scores": { "autonomy": 2, "execution": 3, "verification": 3, "scientificEvidence": 3, "capital": 4, "traction": 3 }, "evidenceLevel": "Commercial partnerships", "accessModel": "Proprietary", "tractionSignals": [ "Daphne Koller founder", "Lilly and Bristol Myers Squibb partnerships", "Reported $2.5B valuation" ], "sourceIds": [ "official-insitro-purpose", "ap-insitro-ai-drug-discovery-202412" ] }, "Cellarity": { "id": "cellarity", "entityType": "Company", "stage": "Private scaleup", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery", "Single-cell biology" ], "stackRoles": [ "Reason", "Data", "Verify" ], "products": [ "Cell State-Correcting therapeutics platform", "Single-cell and ML discovery workflows" ], "scores": { "autonomy": 2, "execution": 2, "verification": 3, "scientificEvidence": 2, "capital": 3, "traction": 2 }, "evidenceLevel": "Platform pipeline", "accessModel": "Proprietary", "tractionSignals": [ "$121M Series C", "$274M total disclosed", "Flagship Pioneering origin" ], "sourceIds": [ "official-cellarity-about", "cellarity-series-c-202210" ] }, "AbCellera": { "id": "abcellera", "entityType": "Public company", "stage": "Public", "region": "North America", "scienceDomains": [ "Biology", "Antibody discovery", "Clinical manufacturing" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Antibody discovery platform", "Single-cell screening", "Clinical manufacturing facility" ], "scores": { "autonomy": 2, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 4, "traction": 4 }, "evidenceLevel": "Public company", "accessModel": "Partnership platform", "tractionSignals": [ "NASDAQ-listed", "$555.5M IPO proceeds", "DARPA pandemic-response platform", "Eli Lilly COVID antibody collaboration" ], "sourceIds": [ "official-abcellera-platform", "renaissance-abcellera-ipo-202011" ] }, "Iambic Therapeutics": { "id": "iambic-therapeutics", "entityType": "Company", "stage": "Clinical-stage platform", "region": "North America", "scienceDomains": [ "Drug discovery", "Chemistry", "Biology" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Enchant", "NeuralPLexer", "AI-driven discovery platform" ], "scores": { "autonomy": 3, "execution": 3, "verification": 3, "scientificEvidence": 3, "capital": 3, "traction": 3 }, "evidenceLevel": "Clinical pipeline", "accessModel": "Proprietary", "tractionSignals": [ "$100M Series B", "NVIDIA collaboration", "Weekly molecular design to biological data loops" ], "sourceIds": [ "official-iambic-platform", "fierce-iambic-series-b-202310" ] }, "Causaly": { "id": "causaly", "entityType": "Company", "stage": "Growth", "region": "Europe", "scienceDomains": [ "Biomedical research", "Drug discovery", "Evidence synthesis" ], "stackRoles": [ "Reason", "Verify", "Data" ], "products": [ "Causaly platform", "Agentic Research", "Biomedical knowledge graph" ], "scores": { "autonomy": 3, "execution": 0, "verification": 3, "scientificEvidence": 2, "capital": 3, "traction": 4 }, "evidenceLevel": "Commercial platform", "accessModel": "Proprietary", "tractionSignals": [ "$60M Series B", "Top pharma customers", "500M facts and 70M directional relationships claimed" ], "sourceIds": [ "official-causaly", "finsmes-causaly-series-b-202307" ] }, "Iktos": { "id": "iktos", "entityType": "Company", "stage": "Commercial platform", "region": "Europe", "scienceDomains": [ "Chemistry", "Drug discovery", "Medicinal chemistry" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Makya", "Spaya", "Iktos Robotics Chemistry", "Iktos Robotics Biology" ], "scores": { "autonomy": 3, "execution": 3, "verification": 2, "scientificEvidence": 3, "capital": 2, "traction": 3 }, "evidenceLevel": "Commercial platform", "accessModel": "Hybrid", "tractionSignals": [ "60+ collaborations", "2024 integrated discovery platform launch", "EIC Accelerator support" ], "sourceIds": [ "official-iktos", "official-iktos-about" ] }, "Genesis Molecular AI": { "id": "genesis-molecular-ai", "entityType": "Company", "stage": "Private scaleup", "region": "North America", "scienceDomains": [ "Drug discovery", "Chemistry", "Molecular AI" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "GEMS", "Pearl", "AI-enabled small-molecule pipeline" ], "scores": { "autonomy": 3, "execution": 2, "verification": 3, "scientificEvidence": 3, "capital": 3, "traction": 3 }, "evidenceLevel": "Platform pipeline", "accessModel": "Proprietary", "tractionSignals": [ "$200M Series B", "Lilly and Genentech collaborations", "San Diego integrated lab" ], "sourceIds": [ "official-genesis-molecular-ai", "fierce-genesis-series-b-202308", "cen-genesis-series-b-202308" ] }, "Argonne National Laboratory": { "id": "argonne-national-laboratory", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Materials", "Chemistry", "Biology", "Energy" ], "stackRoles": [ "Reason", "Execute", "Infrastructure", "Data", "Verify" ], "products": [ "Autonomous Discovery initiative", "Polybot", "Self-driving chemist", "Autonomous discovery data platform" ], "scores": { "autonomy": 4, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Research facility", "accessModel": "Public infrastructure", "tractionSignals": [ "Autonomous Discovery program", "Self-driving chemist project", "Polybot materials work" ], "sourceIds": [ "official-argonne-autonomous-discovery", "official-argonne-science101-autonomous-discovery" ] }, "Oak Ridge National Laboratory": { "id": "oak-ridge-national-laboratory", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Materials", "Chemistry", "Energy", "Computing" ], "stackRoles": [ "Reason", "Execute", "Infrastructure", "Data", "Verify" ], "products": [ "Autonomous Science program", "Autonomous Chemistry Lab", "INTERSECT initiative" ], "scores": { "autonomy": 4, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Research facility", "accessModel": "Public infrastructure", "tractionSignals": [ "Autonomous Science page", "Autonomous Chemistry Lab", "Genesis Mission role" ], "sourceIds": [ "official-ornl-autonomous-science", "official-ornl-autonomous-chemistry-lab" ] }, "Lawrence Livermore National Laboratory": { "id": "lawrence-livermore-national-laboratory", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Materials", "Advanced manufacturing", "AI for science" ], "stackRoles": [ "Reason", "Execute", "Infrastructure", "Data", "Verify" ], "products": [ "APEX self-driving alloy lab", "AI/ML research programs", "Exascale computing infrastructure" ], "scores": { "autonomy": 3, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 2 }, "evidenceLevel": "Research facility", "accessModel": "Public infrastructure", "tractionSignals": [ "APEX platform", "Cornell collaboration", "Planned autonomous alloy workflow" ], "sourceIds": [ "official-llnl-apex-202507", "str-llnl-apex-202601" ] }, "NIST": { "id": "nist", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Materials", "Metrology", "Formulations", "Biology" ], "stackRoles": [ "Execute", "Verify", "Infrastructure", "Data" ], "products": [ "Autonomous laboratories program", "Autonomous Formulation Lab", "CAMEO", "Hermes" ], "scores": { "autonomy": 3, "execution": 4, "verification": 4, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Research facility", "accessModel": "Public infrastructure", "tractionSignals": [ "Autonomous laboratories portfolio", "NIST SP 1320", "Autonomous Formulation Lab" ], "sourceIds": [ "official-nist-autonomous-labs", "nist-autonomous-systems-materials-202009", "nist-autonomous-formulation-lab-202508" ] }, "NREL": { "id": "nrel", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Materials", "Energy", "Characterization" ], "stackRoles": [ "Execute", "Data", "Verify", "Infrastructure" ], "products": [ "Autonomous experimentation program", "Autonomous synthesis", "Autonomous characterization", "AI-generated control software" ], "scores": { "autonomy": 3, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Research facility", "accessModel": "Public infrastructure", "tractionSignals": [ "Autonomous sputter synthesis", "Autonomous characterization", "Digital Discovery control-software paper" ], "sourceIds": [ "official-nrel-autonomous-experimentation", "rsc-text-to-test-202410" ] }, "Brookhaven National Laboratory": { "id": "brookhaven-national-laboratory", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Materials", "Photon science", "Nanoscience" ], "stackRoles": [ "Execute", "Data", "Verify", "Infrastructure" ], "products": [ "NSLS-II autonomous steering tools", "CFN autonomous materials discovery", "Robotics and Bluesky workflows" ], "scores": { "autonomy": 3, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Research facility", "accessModel": "Public infrastructure", "tractionSignals": [ "NSLS-II DSSI robotics work", "AI-discovered nanostructures", "ILLUMINE autonomous experiment steering" ], "sourceIds": [ "official-bnl-nsls2-dssi", "sciencedaily-bnl-ai-nanostructures-202301" ] }, "Materials Genome Initiative": { "id": "materials-genome-initiative", "entityType": "Government initiative", "stage": "Policy and research network", "region": "North America", "scienceDomains": [ "Materials", "Manufacturing", "Policy", "Data infrastructure" ], "stackRoles": [ "Infrastructure", "Data", "Verify" ], "products": [ "Materials Genome Initiative", "Autonomous experimentation reports", "Federal agency coordination" ], "scores": { "autonomy": 1, "execution": 1, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 4 }, "evidenceLevel": "Government initiative", "accessModel": "Public infrastructure", "tractionSignals": [ "2011 launch", "Multi-agency federal coordination", "Autonomous experimentation workshop report" ], "sourceIds": [ "official-mgi-about", "official-mgi-autonomous-experimentation-report" ] }, "Chai Discovery": { "id": "chai-discovery", "entityType": "Company", "stage": "Growth", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery", "Protein design" ], "stackRoles": [ "Reason", "Data", "Verify" ], "products": [ "Chai-1", "Chai-2", "Molecular design platform" ], "scores": { "autonomy": 3, "execution": 1, "verification": 3, "scientificEvidence": 4, "capital": 4, "traction": 4 }, "evidenceLevel": "Wet-lab validated model", "accessModel": "Commercial partnership", "tractionSignals": [ "$130M Series B at $1.3B valuation", "$225M+ total funding", "Eli Lilly biologics collaboration" ], "sourceIds": [ "official-chai-discovery", "businesswire-chai-series-b-202512", "biospace-chai-lilly-202601", "biorxiv-chai2-zero-shot-202507" ] }, "Converge Bio": { "id": "converge-bio", "entityType": "Company", "stage": "Series A", "region": "North America / Middle East", "scienceDomains": [ "Biology", "Drug discovery", "Protein engineering" ], "stackRoles": [ "Reason", "Data", "Verify" ], "products": [ "Target discovery models", "Antibody design system", "Protein manufacturing optimization system" ], "scores": { "autonomy": 2, "execution": 1, "verification": 2, "scientificEvidence": 2, "capital": 3, "traction": 3 }, "evidenceLevel": "Commercial traction", "accessModel": "Commercial partnership", "tractionSignals": [ "$25M Series A", "$30M total raised", "40+ pharma and biotech programs" ], "sourceIds": [ "official-converge-series-a-202601", "techcrunch-converge-series-a-202601" ] }, "Generare": { "id": "generare", "entityType": "Company", "stage": "Series A", "region": "Europe", "scienceDomains": [ "Natural products", "Drug discovery", "Synthetic biology" ], "stackRoles": [ "Data", "Execute", "Reason" ], "products": [ "Microbial genome-screening platform", "Cryptic small-molecule dataset", "Synthetic-biology expression workflow" ], "scores": { "autonomy": 2, "execution": 3, "verification": 2, "scientificEvidence": 3, "capital": 3, "traction": 3 }, "evidenceLevel": "Commercial data-generation platform", "accessModel": "Commercial partnership", "tractionSignals": [ "€20M Series A", "200 previously uncharacterized small molecules", "10x dataset scale target by 2027" ], "sourceIds": [ "alven-generare-series-a-202604", "biologydigital-generare-series-a-202604" ] }, "Alloy Therapeutics": { "id": "alloy-therapeutics", "entityType": "Company", "stage": "Growth", "region": "North America", "scienceDomains": [ "Biologics", "Drug discovery", "Biotech infrastructure" ], "stackRoles": [ "Execute", "Data", "Infrastructure" ], "products": [ "Biologics discovery platforms", "Integrated preclinical services", "AI and real-world-data infrastructure" ], "scores": { "autonomy": 1, "execution": 3, "verification": 2, "scientificEvidence": 2, "capital": 3, "traction": 4 }, "evidenceLevel": "Commercial infrastructure", "accessModel": "Commercial partnership", "tractionSignals": [ "$40M Series E at $1B valuation", "200+ partners", "22 clinical programs" ], "sourceIds": [ "businesswire-alloy-series-e-202604", "citybiz-alloy-series-e-202604" ] }, "RoboChem-Flex": { "id": "robochem-flex", "entityType": "Academic platform", "stage": "Peer-reviewed prototype", "region": "Europe", "scienceDomains": [ "Chemistry", "Lab automation", "Optimization" ], "stackRoles": [ "Execute", "Reason", "Infrastructure", "Verify" ], "products": [ "RoboChem-Flex modular SDL", "Python control framework", "Bayesian optimization workflow" ], "scores": { "autonomy": 3, "execution": 3, "verification": 3, "scientificEvidence": 4, "capital": 1, "traction": 2 }, "evidenceLevel": "Peer-reviewed prototype", "accessModel": "Open research", "tractionSignals": [ "Nature Synthesis 2026", "Six validation case studies", "Low-cost modular hardware" ], "sourceIds": [ "nature-robochem-flex-202604" ] }, "Scivity Labs": { "id": "scivity-labs", "entityType": "Company", "stage": "Pre-seed", "region": "Europe", "scienceDomains": [ "ML research", "Biology", "Chemistry", "General science" ], "stackRoles": [ "Reason", "Execute", "Verify", "Infrastructure" ], "products": [ "Autonomous Laboratory Platform", "Verification surface", "AI for Science Landscape" ], "scores": { "autonomy": 3, "execution": 2, "verification": 2, "scientificEvidence": 1, "capital": 1, "traction": 1 }, "scoreNotes": { "verification": "Conservative score: public materials describe validation outcomes while thresholds, calibration, and check-composition logic remain private." }, "evidenceLevel": "Self-reported platform", "accessModel": "Proprietary", "tractionSignals": [ "Pre-seed company", "Public verification surface", "First autonomous end-to-end research run disclosed" ], "sourceIds": [ "official-scivity-home-202605", "official-scivity-verification-202603", "official-scivity-changelog-202604" ] }, "Tetsuwan Scientific": { "id": "tetsuwan-scientific", "entityType": "Company", "stage": "Pre-seed", "region": "North America", "scienceDomains": [ "Biology", "Lab automation", "Cloud lab" ], "stackRoles": [ "Reason", "Execute", "Infrastructure" ], "products": [ "ResearchOS", "Tetsuwan Cloud Lab", "LabAutoWiki" ], "scores": { "autonomy": 2, "execution": 3, "verification": 1, "scientificEvidence": 1, "capital": 1, "traction": 2 }, "evidenceLevel": "Pilot", "accessModel": "Proprietary", "tractionSignals": [ "$2.7M pre-seed", "First lab deployment reported", "ResearchOS execution-layer positioning" ], "sourceIds": [ "official-tetsuwan-home-2026", "techcrunch-tetsuwan-preseed-202412", "synbiobeta-tetsuwan-preseed-202411" ] }, "Microsoft Discovery": { "id": "microsoft-discovery", "entityType": "Platform initiative", "stage": "Preview", "region": "North America", "scienceDomains": [ "Biology", "Chemistry", "Materials", "Semiconductors" ], "stackRoles": [ "Reason", "Execute", "Verify", "Data", "Infrastructure" ], "products": [ "Microsoft Discovery Engine", "Azure-governed R&D platform", "Agentic experiment loop" ], "scores": { "autonomy": 3, "execution": 2, "verification": 3, "scientificEvidence": 2, "capital": 4, "traction": 2 }, "evidenceLevel": "Enterprise preview", "accessModel": "Preview / enterprise", "tractionSignals": [ "Build 2025 launch", "Expanded preview in 2026", "Azure-native governance and HPC" ], "sourceIds": [ "official-microsoft-discovery-solution-202604", "official-microsoft-discovery-intro-202505", "official-microsoft-discovery-preview-202604", "techcrunch-microsoft-discovery-202505" ] }, "Anthropic": { "id": "anthropic", "entityType": "Frontier AI lab", "stage": "Frontier platform", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery", "General science", "Reasoning" ], "stackRoles": [ "Reason", "Verify", "Data" ], "products": [ "Claude family", "Claude for Life Sciences", "BioMysteryBench", "Agent Skills", "Coefficient Bio team (acquired)" ], "scores": { "autonomy": 3, "execution": 1, "verification": 3, "scientificEvidence": 3, "capital": 4, "traction": 4 }, "evidenceLevel": "Frontier platform", "accessModel": "Proprietary", "tractionSignals": [ "Top of AstaBench scientific-agent leaderboard", "Allen Institute and HHMI founding partnerships", "Coefficient Bio acquisition", "Claude for Life Sciences launch" ], "sourceIds": [ "official-anthropic-claude-life-sciences-202510", "official-anthropic-allen-hhmi-202510", "official-anthropic-biomysterybench-202604", "techcrunch-anthropic-coefficient-bio-202604", "biospace-anthropic-coefficient-202604", "ai2-astabench-update-202604" ] }, "Ai2": { "id": "ai2", "entityType": "Nonprofit research institute", "stage": "Research platform", "region": "North America", "scienceDomains": [ "General science", "Literature", "Biology", "Reasoning" ], "stackRoles": [ "Reason", "Verify", "Data" ], "products": [ "Asta agents", "AstaBench", "Scientific Corpus Tool", "OLMo models", "Semantic Scholar" ], "scores": { "autonomy": 3, "execution": 1, "verification": 4, "scientificEvidence": 4, "capital": 3, "traction": 3 }, "evidenceLevel": "Open research", "accessModel": "Open", "tractionSignals": [ "AstaBench ICLR 2026 oral", "UK AISI and General Reasoning adoption", "Submissions from Elicit, SciSpace, Distyl AI, EvoScientist", "Asta runs on 108M+ abstracts and 200M+ paper corpus" ], "sourceIds": [ "official-ai2-asta-launch-202508", "official-ai2-asta-platform", "official-astabench", "ai2-astabench-blog-202508", "ai2-astabench-update-202604", "businesswire-ai2-asta-202508" ] }, "Amazon Bio Discovery": { "id": "amazon-bio-discovery", "entityType": "Platform initiative", "stage": "Preview", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery" ], "stackRoles": [ "Reason", "Execute", "Data", "Infrastructure" ], "products": [ "Amazon Bio Discovery", "Biology foundation model catalog", "Lab-in-the-loop orchestration", "Bedrock agents" ], "scores": { "autonomy": 3, "execution": 2, "verification": 2, "scientificEvidence": 2, "capital": 4, "traction": 2 }, "evidenceLevel": "Enterprise preview", "accessModel": "Preview / enterprise", "tractionSignals": [ "April 2026 launch with 40+ BioFMs", "Ginkgo Nebula and Twist Bioscience integration", "Memorial Sloan Kettering early project", "Closed lab-in-the-loop on AWS" ], "sourceIds": [ "official-aws-bio-discovery-blog-202604", "genengnews-aws-bio-discovery-202604", "aipedia-aws-bio-discovery-202604", "official-ginkgo-q1-2026" ] }, "Profluent": { "id": "profluent", "entityType": "Company", "stage": "Growth", "region": "North America", "scienceDomains": [ "Biology", "Protein design", "Gene editing" ], "stackRoles": [ "Reason", "Data", "Verify" ], "products": [ "Protein foundation models", "OpenCRISPR-1", "ProGen", "Site-specific recombinase platform" ], "scores": { "autonomy": 2, "execution": 2, "verification": 3, "scientificEvidence": 4, "capital": 4, "traction": 4 }, "evidenceLevel": "Wet-lab validated model", "accessModel": "Hybrid", "tractionSignals": [ "$106M Series B led by Altimeter and Bezos", "$2.25B Lilly recombinase pact", "OpenCRISPR-1 open release", "Nature 2025 AI-designed CRISPR system" ], "sourceIds": [ "official-profluent", "businesswire-profluent-series-b-202511", "biospace-profluent-lilly-202604", "airstreet-profluent-lilly-202604" ] } }, "entity-details-by-id": { "periodic-labs": { "id": "periodic-labs", "entityType": "Company", "stage": "Private scaleup", "region": "North America", "scienceDomains": [ "Materials", "Chemistry" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Autonomous materials labs", "AI scientist workflow", "Experimental data flywheel" ], "scores": { "autonomy": 4, "execution": 4, "verification": 3, "scientificEvidence": 3, "capital": 4, "traction": 2 }, "evidenceLevel": "Announced platform", "accessModel": "Proprietary", "tractionSignals": [ "$300M seed", "Frontier-lab founding team", "Materials-first autonomous lab buildout" ], "sourceIds": [ "official-periodic-labs" ] }, "lila-sciences": { "id": "lila-sciences", "entityType": "Company", "stage": "Private scaleup", "region": "North America", "scienceDomains": [ "Biology", "Chemistry", "Materials" ], "stackRoles": [ "Reason", "Execute", "Verify", "Data" ], "products": [ "AI Science Factories", "Reasoning models", "Robotic lab network" ], "scores": { "autonomy": 4, "execution": 4, "verification": 3, "scientificEvidence": 3, "capital": 4, "traction": 2 }, "evidenceLevel": "Announced platform", "accessModel": "Proprietary", "tractionSignals": [ "$550M raised", "Flagship Pioneering origin", "Life, chemical, and materials scope" ], "sourceIds": [ "official-lila-sciences" ] }, "sakana-ai": { "id": "sakana-ai", "entityType": "Company", "stage": "Private scaleup", "region": "Asia-Pacific", "scienceDomains": [ "ML research", "General science" ], "stackRoles": [ "Reason", "Verify" ], "products": [ "AI Scientist", "AI Scientist-v2", "Marlin" ], "scores": { "autonomy": 4, "execution": 1, "verification": 2, "scientificEvidence": 3, "capital": 4, "traction": 3 }, "evidenceLevel": "Peer-reviewed demo", "accessModel": "Hybrid", "tractionSignals": [ "AI-authored paper workflow", "$379M cumulative funding", "Commercial enterprise research beta" ], "sourceIds": [ "official-sakana-ai" ] }, "futurehouse": { "id": "futurehouse", "entityType": "Nonprofit lab", "stage": "Research platform", "region": "North America", "scienceDomains": [ "Biology", "Literature", "General science" ], "stackRoles": [ "Reason", "Execute", "Verify", "Data" ], "products": [ "Crow", "Falcon", "Owl", "Phoenix", "PaperQA2", "Robin", "ether0", "DISCO" ], "scores": { "autonomy": 4, "execution": 1, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Open research", "accessModel": "Hybrid", "tractionSignals": [ "Open scientific agents", "LitQA2 performance", "DISCO wet-lab validation", "Commercial spinout via Edison Scientific" ], "sourceIds": [ "official-futurehouse", "official-futurehouse-disco-202604" ] }, "autoscience-institute": { "id": "autoscience-institute", "entityType": "Company", "stage": "Seed", "region": "North America", "scienceDomains": [ "ML research" ], "stackRoles": [ "Reason", "Verify" ], "products": [ "Carl", "Mira" ], "scores": { "autonomy": 4, "execution": 1, "verification": 2, "scientificEvidence": 3, "capital": 2, "traction": 2 }, "evidenceLevel": "Peer-reviewed demo", "accessModel": "Proprietary", "tractionSignals": [ "$14M seed", "ICLR workshop acceptances", "Live Kaggle competition entries" ], "sourceIds": [ "official-autoscience-institute" ] }, "thesis": { "id": "thesis", "entityType": "Company", "stage": "Seed", "region": "North America", "scienceDomains": [ "ML research" ], "stackRoles": [ "Reason", "Verify" ], "products": [ "Automated ML research workflow" ], "scores": { "autonomy": 3, "execution": 1, "verification": 2, "scientificEvidence": 2, "capital": 1, "traction": 1 }, "evidenceLevel": "Benchmark result", "accessModel": "Proprietary", "tractionSignals": [ "YC F25", "MLE-Bench medal-rate claim", "Stanford AI Lab founding network" ], "sourceIds": [ "official-thesis" ] }, "zeon-systems": { "id": "zeon-systems", "entityType": "Company", "stage": "Seed", "region": "North America", "scienceDomains": [ "Lab automation", "Biology" ], "stackRoles": [ "Execute", "Infrastructure" ], "products": [ "Natural-language lab robotics layer", "Labware vision system" ], "scores": { "autonomy": 2, "execution": 3, "verification": 1, "scientificEvidence": 1, "capital": 1, "traction": 1 }, "evidenceLevel": "Pilot", "accessModel": "Proprietary", "tractionSignals": [ "YC X25", "UCSF and Stanford pilots", "Robot-arm protocol execution" ], "sourceIds": [ "official-zeon-systems" ] }, "automata": { "id": "automata", "entityType": "Company", "stage": "Growth", "region": "Europe", "scienceDomains": [ "Biology", "Diagnostics", "Lab automation" ], "stackRoles": [ "Execute", "Infrastructure", "Data" ], "products": [ "LINQ", "Automated lab benches", "LINQ Cloud" ], "scores": { "autonomy": 2, "execution": 4, "verification": 2, "scientificEvidence": 2, "capital": 2, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "Five top pharma customers", "$45M Series C", "Danaher strategic participation" ], "sourceIds": [ "official-automata" ] }, "cenevo": { "id": "cenevo", "entityType": "Company", "stage": "Commercial", "region": "North America / Europe", "scienceDomains": [ "Biology", "Lab automation", "Scientific informatics" ], "stackRoles": [ "Execute", "Data", "Infrastructure" ], "products": [ "Labguru ELN/LIMS", "Titian Mosaic sample management", "AI Protocol Conversion", "AI Automation agent" ], "scores": { "autonomy": 2, "execution": 3, "verification": 2, "scientificEvidence": 2, "capital": 2, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "950+ customer organizations", "45,000+ users", "8 of the top 10 pharma companies" ], "sourceIds": [ "official-cenevo-rebrand-202507", "rdworld-cenevo-ai-agents-202602" ] }, "emerald-cloud-lab": { "id": "emerald-cloud-lab", "entityType": "Company", "stage": "Commercial", "region": "North America", "scienceDomains": [ "Chemistry", "Biology", "Cloud lab" ], "stackRoles": [ "Execute", "Infrastructure", "Data", "Verify" ], "products": [ "Remote programmable lab", "Scientific computing environment" ], "scores": { "autonomy": 2, "execution": 4, "verification": 3, "scientificEvidence": 3, "capital": 2, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "Mature programmable cloud lab", "Academic and pharma users", "Remote experiment execution" ], "sourceIds": [ "official-emerald-cloud-lab" ] }, "strateos": { "id": "strateos", "entityType": "Company", "stage": "Commercial", "region": "North America", "scienceDomains": [ "Biology", "Chemistry", "Cloud lab" ], "stackRoles": [ "Execute", "Infrastructure", "Data" ], "products": [ "Robotic cloud lab", "API experiment orchestration", "On-prem deployments" ], "scores": { "autonomy": 2, "execution": 4, "verification": 2, "scientificEvidence": 2, "capital": 2, "traction": 3 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "Pharma R&D focus", "$56M funding", "API-first orchestration" ], "sourceIds": [ "official-strateos" ] }, "chemify": { "id": "chemify", "entityType": "Company", "stage": "Growth", "region": "Europe", "scienceDomains": [ "Chemistry", "Drug discovery", "Programmable synthesis" ], "stackRoles": [ "Reason", "Execute", "Verify", "Infrastructure" ], "products": [ "Chemputer", "Chemifarm", "Reaction planning software" ], "scores": { "autonomy": 3, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 3, "traction": 3 }, "evidenceLevel": "Published platform", "accessModel": "Proprietary", "tractionSignals": [ "Chemistry-as-code pedigree", "Series B expansion", "Peer-reviewed validation" ], "sourceIds": [ "official-chemify" ] }, "recursion-pharmaceuticals": { "id": "recursion-pharmaceuticals", "entityType": "Public company", "stage": "Public", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery", "Imaging" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Recursion OS", "Industrial phenomics platform", "BioHive supercomputer" ], "scores": { "autonomy": 2, "execution": 3, "verification": 3, "scientificEvidence": 4, "capital": 4, "traction": 4 }, "evidenceLevel": "Public company", "accessModel": "Proprietary", "tractionSignals": [ "NASDAQ-listed", "Exscientia merger", "Industrial-scale image-based screening" ], "sourceIds": [ "official-recursion-pharmaceuticals" ] }, "atinary": { "id": "atinary", "entityType": "Company", "stage": "Commercial", "region": "Europe", "scienceDomains": [ "Chemistry", "Materials", "Drug discovery" ], "stackRoles": [ "Reason", "Execute", "Verify", "Data" ], "products": [ "Self-driving lab software", "Boston Scientific Discovery Factories", "Bayesian optimization workflows" ], "scores": { "autonomy": 3, "execution": 3, "verification": 3, "scientificEvidence": 3, "capital": 1, "traction": 2 }, "evidenceLevel": "Commercial hardware and software", "accessModel": "Proprietary", "tractionSignals": [ "Software-first self-driving lab approach", "Boston Scientific Discovery Factories announced in 2026", "Chemistry informatics partnerships", "Enterprise R&D focus" ], "sourceIds": [ "official-atinary", "official-atinary-boston-sdl-202602", "selectscience-atinary-boston-sdl-202602" ] }, "ginkgo-bioworks": { "id": "ginkgo-bioworks", "entityType": "Public company", "stage": "Public", "region": "North America", "scienceDomains": [ "Biology", "Synthetic biology", "Cloud lab" ], "stackRoles": [ "Reason", "Execute", "Data", "Infrastructure" ], "products": [ "Foundry", "Nebula", "Ginkgo Cloud Lab", "EstiMate" ], "scores": { "autonomy": 3, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 3, "traction": 4 }, "evidenceLevel": "Public company", "accessModel": "Hybrid", "tractionSignals": [ "NYSE-listed", "Nebula autonomous lab", "Large biological data generation platform" ], "sourceIds": [ "official-ginkgo-bioworks" ] }, "radical-ai": { "id": "radical-ai", "entityType": "Company", "stage": "Seed", "region": "North America", "scienceDomains": [ "Materials", "Simulation" ], "stackRoles": [ "Reason", "Execute", "Data" ], "products": [ "Materials foundation models", "TorchSim", "Autonomous materials lab" ], "scores": { "autonomy": 3, "execution": 3, "verification": 2, "scientificEvidence": 2, "capital": 2, "traction": 2 }, "evidenceLevel": "Open research", "accessModel": "Hybrid", "tractionSignals": [ "TorchSim open-source release", "AFWERX contract", "Brooklyn autonomous materials lab" ], "sourceIds": [ "official-radical-ai" ] }, "xaira-therapeutics": { "id": "xaira-therapeutics", "entityType": "Company", "stage": "Private scaleup", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery", "Virtual cell" ], "stackRoles": [ "Reason", "Data", "Verify" ], "products": [ "X-Atlas", "X-Cell", "AI drug discovery pipeline" ], "scores": { "autonomy": 2, "execution": 2, "verification": 3, "scientificEvidence": 3, "capital": 4, "traction": 2 }, "evidenceLevel": "Published data asset", "accessModel": "Proprietary", "tractionSignals": [ "$1B+ Series A", "25.6M-cell atlas", "Nobel laureate founder network" ], "sourceIds": [ "official-xaira-therapeutics" ] }, "helical": { "id": "helical", "entityType": "Company", "stage": "Seed", "region": "Europe", "scienceDomains": [ "Biology", "Drug discovery", "Biomarkers" ], "stackRoles": [ "Reason", "Data", "Verify" ], "products": [ "Virtual Lab", "Model Factory", "Bio foundation model workflows" ], "scores": { "autonomy": 3, "execution": 1, "verification": 3, "scientificEvidence": 2, "capital": 2, "traction": 3 }, "evidenceLevel": "Production deployment", "accessModel": "Proprietary", "tractionSignals": [ "$10M seed", "Production deployments with multiple top-20 pharma companies", "Public Pfizer collaboration on predictive blood-based safety biomarkers" ], "sourceIds": [ "official-helical", "globenewswire-helical-seed-202604", "techeu-helical-seed-202604" ] }, "fathom-therapeutics": { "id": "fathom-therapeutics", "entityType": "Company", "stage": "Series A", "region": "North America", "scienceDomains": [ "Drug discovery", "Chemistry", "Biology" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Microcosmos", "AI-enabled small-molecule discovery engine", "Lab-in-the-loop discovery programs" ], "scores": { "autonomy": 2, "execution": 2, "verification": 2, "scientificEvidence": 3, "capital": 3, "traction": 3 }, "evidenceLevel": "Working prototype", "accessModel": "Proprietary", "tractionSignals": [ "$47M Series A", "Lead program in animal efficacy studies", "Multiple external discovery partnerships" ], "sourceIds": [ "official-fathom-therapeutics", "prnewswire-fathom-series-a-202604", "cen-fathom-series-a-202604" ] }, "10x-science": { "id": "10x-science", "entityType": "Company", "stage": "Seed", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery", "Protein characterization" ], "stackRoles": [ "Data", "Verify" ], "products": [ "AI-native protein characterization platform", "Mass spectrometry analysis agents", "Proteoform analysis workflows" ], "scores": { "autonomy": 2, "execution": 1, "verification": 3, "scientificEvidence": 2, "capital": 1, "traction": 2 }, "evidenceLevel": "Announced platform", "accessModel": "Proprietary", "tractionSignals": [ "$4.8M seed", "YC W26", "Stanford/Bertozzi-lab founder expertise" ], "sourceIds": [ "official-10x-science", "yc-10x-science-profile-202604", "techcrunch-10x-science-seed-202604", "prnewswire-10x-science-seed-202604" ] }, "a-lab-lbnl": { "id": "a-lab-lbnl", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Materials", "Chemistry" ], "stackRoles": [ "Reason", "Execute", "Verify", "Data" ], "products": [ "Autonomous materials synthesis facility" ], "scores": { "autonomy": 4, "execution": 4, "verification": 2, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Published platform", "accessModel": "Public infrastructure", "tractionSignals": [ "Nature-published autonomous synthesis", "DOE national lab context", "Defining verification case study" ], "sourceIds": [ "official-a-lab-lbnl" ] }, "acceleration-consortium": { "id": "acceleration-consortium", "entityType": "Academic network", "stage": "Research network", "region": "North America", "scienceDomains": [ "Materials", "Chemistry", "Standards" ], "stackRoles": [ "Execute", "Verify", "Infrastructure", "Data" ], "products": [ "Self-driving lab network", "Protocols", "Standards coordination" ], "scores": { "autonomy": 2, "execution": 3, "verification": 3, "scientificEvidence": 4, "capital": 3, "traction": 3 }, "evidenceLevel": "Research network", "accessModel": "Public infrastructure", "tractionSignals": [ "$200M CFREF grant", "30+ member institutions", "Standards role" ], "sourceIds": [ "official-acceleration-consortium" ] }, "pnnl": { "id": "pnnl", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Chemistry", "Energy", "Materials" ], "stackRoles": [ "Execute", "Verify", "Infrastructure", "Data" ], "products": [ "Autonomy Studio", "DOE autonomous experimentation programs" ], "scores": { "autonomy": 2, "execution": 3, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Research facility", "accessModel": "Public infrastructure", "tractionSignals": [ "DOE national lab", "Autonomy Studio", "Genesis Mission network" ], "sourceIds": [ "official-pnnl" ] }, "google-deepmind": { "id": "google-deepmind", "entityType": "Frontier AI lab", "stage": "Big Tech lab", "region": "Europe", "scienceDomains": [ "Biology", "Materials", "Mathematics", "General science" ], "stackRoles": [ "Reason", "Verify", "Data" ], "products": [ "AlphaFold", "AlphaProteo", "GNoME", "AlphaGeometry", "AI co-scientist" ], "scores": { "autonomy": 3, "execution": 1, "verification": 3, "scientificEvidence": 4, "capital": 4, "traction": 4 }, "evidenceLevel": "Frontier research", "accessModel": "Hybrid", "tractionSignals": [ "Nobel-recognized AlphaFold work", "DOE Genesis Mission partnership", "Deep science AI bench" ], "sourceIds": [ "official-google-deepmind", "epoch-capabilities" ] }, "openai": { "id": "openai", "entityType": "Frontier AI lab", "stage": "Frontier platform", "region": "North America", "scienceDomains": [ "General science", "Biology", "Reasoning" ], "stackRoles": [ "Reason", "Verify", "Infrastructure" ], "products": [ "Deep Research", "o-series reasoning models", "GPT-Rosalind", "Prism" ], "scores": { "autonomy": 3, "execution": 1, "verification": 3, "scientificEvidence": 3, "capital": 4, "traction": 4 }, "evidenceLevel": "Frontier platform", "accessModel": "Proprietary", "tractionSignals": [ "DOE national labs access", "Science policy agenda", "Specialized biology model launch" ], "sourceIds": [ "official-openai", "artificial-analysis-leaderboard", "artificial-analysis-methodology" ] }, "isomorphic-labs": { "id": "isomorphic-labs", "entityType": "Company", "stage": "Private scaleup", "region": "Europe", "scienceDomains": [ "Biology", "Drug discovery" ], "stackRoles": [ "Reason", "Verify", "Data" ], "products": [ "IsoDDE", "AI drug design platform" ], "scores": { "autonomy": 2, "execution": 2, "verification": 3, "scientificEvidence": 4, "capital": 4, "traction": 3 }, "evidenceLevel": "Clinical pipeline", "accessModel": "Proprietary", "tractionSignals": [ "$600M first external round", "Nearly $3B in pharma collaborations", "17-program pipeline" ], "sourceIds": [ "official-isomorphic-labs" ] }, "edison-scientific": { "id": "edison-scientific", "entityType": "Company", "stage": "Seed", "region": "North America", "scienceDomains": [ "Biology", "Materials", "Statistical genetics", "General science" ], "stackRoles": [ "Reason", "Verify", "Data" ], "products": [ "Kosmos", "Edison Literature", "PaperQA3", "LABBench2" ], "scores": { "autonomy": 4, "execution": 1, "verification": 3, "scientificEvidence": 3, "capital": 3, "traction": 2 }, "evidenceLevel": "Launch paper", "accessModel": "Proprietary", "tractionSignals": [ "$70M seed", "Seven demonstrated discoveries", "PaperQA3 literature upgrade", "NVIDIA scaling partnership", "FutureHouse commercial spinout" ], "sourceIds": [ "official-edison-scientific", "official-edison-paperqa3-202602", "official-edison-nvidia-202603", "official-edison-labbench2-202605" ] }, "insilico-medicine": { "id": "insilico-medicine", "entityType": "Public company", "stage": "Public", "region": "Asia-Pacific", "scienceDomains": [ "Biology", "Drug discovery", "Clinical development" ], "stackRoles": [ "Reason", "Verify", "Data" ], "products": [ "Pharma.AI", "rentosertib", "AI drug discovery pipeline" ], "scores": { "autonomy": 2, "execution": 2, "verification": 3, "scientificEvidence": 4, "capital": 4, "traction": 4 }, "evidenceLevel": "Clinical data", "accessModel": "Proprietary", "tractionSignals": [ "HKEX listing", "Phase 2a clinical data", "$2.75B Lilly partnership" ], "sourceIds": [ "official-insilico-medicine" ] }, "ami-labs": { "id": "ami-labs", "entityType": "Frontier AI lab", "stage": "Seed", "region": "Europe", "scienceDomains": [ "World models", "General intelligence", "Physical reasoning" ], "stackRoles": [ "Reason", "Data" ], "products": [ "JEPA-first world model research agenda" ], "scores": { "autonomy": 2, "execution": 1, "verification": 1, "scientificEvidence": 1, "capital": 4, "traction": 1 }, "evidenceLevel": "Announced lab", "accessModel": "Proprietary", "tractionSignals": [ "$1.03B seed", "Yann LeCun founding team", "World-model-first thesis" ], "sourceIds": [ "official-ami-labs" ] }, "chemlex": { "id": "chemlex", "entityType": "Company", "stage": "Growth", "region": "Asia-Pacific", "scienceDomains": [ "Chemistry", "Drug discovery", "Programmable synthesis" ], "stackRoles": [ "Reason", "Execute", "Infrastructure", "Data" ], "products": [ "24/7 autonomous synthesis line", "Chemistry CRO workflow" ], "scores": { "autonomy": 3, "execution": 4, "verification": 2, "scientificEvidence": 2, "capital": 3, "traction": 3 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "70+ customers", "Six top-ten pharma customers", "Singapore flagship self-driving lab" ], "sourceIds": [ "official-chemlex" ] }, "coscientist-cmu": { "id": "coscientist-cmu", "entityType": "Academic project", "stage": "Published research", "region": "North America", "scienceDomains": [ "Chemistry", "Autonomous agents", "Cloud lab" ], "stackRoles": [ "Reason", "Execute", "Verify", "Data" ], "products": [ "Coscientist autonomous chemistry system", "Cloud-lab experiment execution" ], "scores": { "autonomy": 4, "execution": 3, "verification": 3, "scientificEvidence": 4, "capital": 1, "traction": 2 }, "evidenceLevel": "Nature paper", "accessModel": "Research prototype", "tractionSignals": [ "Nature 2023 publication", "Emerald Cloud Lab execution", "Palladium cross-coupling demonstrations" ], "sourceIds": [ "cmu-coscientist-202312", "nature-coscientist-202312" ] }, "kebotix": { "id": "kebotix", "entityType": "Company", "stage": "Commercial", "region": "North America", "scienceDomains": [ "Materials", "Chemistry", "Green chemistry" ], "stackRoles": [ "Reason", "Execute", "Data" ], "products": [ "Self-driving materials discovery platform", "AI-guided chemistry workflows" ], "scores": { "autonomy": 3, "execution": 3, "verification": 2, "scientificEvidence": 2, "capital": 2, "traction": 2 }, "evidenceLevel": "Commercial platform", "accessModel": "Proprietary", "tractionSignals": [ "$11.4M Series A", "Novo Holdings lead", "Self-driving-lab positioning" ], "sourceIds": [ "official-kebotix", "finsmes-kebotix-series-a-202004" ] }, "citrine-informatics": { "id": "citrine-informatics", "entityType": "Company", "stage": "Growth", "region": "North America", "scienceDomains": [ "Materials", "Chemistry", "Manufacturing" ], "stackRoles": [ "Reason", "Data", "Verify" ], "products": [ "Citrine Platform", "Materials AI models", "R&D data infrastructure" ], "scores": { "autonomy": 2, "execution": 1, "verification": 3, "scientificEvidence": 3, "capital": 3, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "$16M Series C", "Global chemicals customers", "Materials AI operating-platform positioning" ], "sourceIds": [ "official-citrine-informatics", "businesswire-citrine-series-c-202301" ] }, "artificial": { "id": "artificial", "entityType": "Company", "stage": "Series A", "region": "North America", "scienceDomains": [ "Lab automation", "Biology", "Synthetic biology" ], "stackRoles": [ "Execute", "Infrastructure", "Data" ], "products": [ "aLab Suite", "Lab orchestration platform", "Digital twin builder" ], "scores": { "autonomy": 3, "execution": 3, "verification": 2, "scientificEvidence": 2, "capital": 2, "traction": 2 }, "evidenceLevel": "Commercial platform", "accessModel": "Proprietary", "tractionSignals": [ "$21.5M Series A", "Thermo Fisher and Beam Therapeutics customers", "AI agents for lab process optimization" ], "sourceIds": [ "official-artificial", "techcrunch-artificial-series-a-202105" ] }, "synthace": { "id": "synthace", "entityType": "Company", "stage": "Growth", "region": "Europe", "scienceDomains": [ "Biology", "Experiment design", "Lab automation" ], "stackRoles": [ "Reason", "Execute", "Data" ], "products": [ "Synthace platform", "Automated multivariate experiments", "Automation instruction generation" ], "scores": { "autonomy": 2, "execution": 3, "verification": 2, "scientificEvidence": 2, "capital": 3, "traction": 3 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "$35M Series C", "Top pharma and CRO customers", "No-code DOE-to-automation workflow" ], "sourceIds": [ "official-synthace", "finsmes-synthace-series-c-202111" ] }, "tetrascience": { "id": "tetrascience", "entityType": "Company", "stage": "Growth", "region": "North America", "scienceDomains": [ "Scientific data", "Biopharma", "Lab automation" ], "stackRoles": [ "Data", "Infrastructure", "Verify" ], "products": [ "Tetra Scientific Data and AI Platform", "Tetra Data Platform", "Tetra AI" ], "scores": { "autonomy": 2, "execution": 1, "verification": 3, "scientificEvidence": 2, "capital": 3, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "$80M Series B", "Enterprise pharma adoption", "Instrument and informatics partner network" ], "sourceIds": [ "official-tetrascience", "fierce-tetrascience-series-b-202104" ] }, "riffyn": { "id": "riffyn", "entityType": "Company / product line", "stage": "Commercial", "region": "North America", "scienceDomains": [ "Bioprocessing", "Scientific data", "Experiment design" ], "stackRoles": [ "Data", "Verify", "Infrastructure" ], "products": [ "Riffyn X", "Scientific Development Environment", "Process data models" ], "scores": { "autonomy": 1, "execution": 1, "verification": 3, "scientificEvidence": 2, "capital": 2, "traction": 3 }, "evidenceLevel": "Commercial", "accessModel": "Proprietary", "tractionSignals": [ "$15M Series B", "BASF collaboration", "Siemens / Dotmatics current product line" ], "sourceIds": [ "siemens-riffyn-x", "biospace-riffyn-series-b-201905" ] }, "opentrons": { "id": "opentrons", "entityType": "Company", "stage": "Commercial", "region": "North America", "scienceDomains": [ "Lab automation", "Biology", "Liquid handling" ], "stackRoles": [ "Execute", "Infrastructure" ], "products": [ "Opentrons Flex", "OT-2", "Python Protocol API", "Protocol Designer" ], "scores": { "autonomy": 2, "execution": 4, "verification": 2, "scientificEvidence": 3, "capital": 2, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Hybrid", "tractionSignals": [ "Open-source software and firmware", "Flex launch", "Mayo Clinic, Harvard, and MIT customer signals" ], "sourceIds": [ "official-opentrons-flex", "techcrunch-opentrons-flex-202305" ] }, "tecan": { "id": "tecan", "entityType": "Public company", "stage": "Public", "region": "Europe", "scienceDomains": [ "Lab automation", "Diagnostics", "Life sciences" ], "stackRoles": [ "Execute", "Infrastructure" ], "products": [ "Fluent", "Freedom EVO", "Tecan Labwerx", "Liquid handling systems" ], "scores": { "autonomy": 1, "execution": 4, "verification": 2, "scientificEvidence": 3, "capital": 3, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Commercial hardware", "tractionSignals": [ "SIX-listed", "3,000+ employees", "Global lab automation installed base" ], "sourceIds": [ "official-tecan-history", "stockanalysis-tecan-profile-202603" ] }, "highres-biosolutions": { "id": "highres-biosolutions", "entityType": "Company", "stage": "Commercial", "region": "North America", "scienceDomains": [ "Lab automation", "Drug discovery", "Synthetic biology", "Genomics" ], "stackRoles": [ "Execute", "Infrastructure", "Data" ], "products": [ "Cellario", "Nucleus", "Prime liquid handler", "AutoPod" ], "scores": { "autonomy": 2, "execution": 4, "verification": 2, "scientificEvidence": 2, "capital": 2, "traction": 4 }, "evidenceLevel": "Commercial", "accessModel": "Commercial hardware and software", "tractionSignals": [ "20+ years deployments", "Axel Johnson ownership", "Cellario OS orchestration platform" ], "sourceIds": [ "official-highres", "biodev-highres-cellario-202509" ] }, "generate-biomedicines": { "id": "generate-biomedicines", "entityType": "Company", "stage": "Clinical-stage platform", "region": "North America", "scienceDomains": [ "Biology", "Protein design", "Drug discovery" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Generate Platform", "Generative Biology", "Protein therapeutic pipeline" ], "scores": { "autonomy": 2, "execution": 3, "verification": 3, "scientificEvidence": 3, "capital": 4, "traction": 3 }, "evidenceLevel": "Clinical pipeline", "accessModel": "Proprietary", "tractionSignals": [ "$370M Series B", "$273M Series C", "Amgen collaboration", "17-program pipeline reported in 2023" ], "sourceIds": [ "official-generate-biomedicines", "fierce-generate-series-b-202111", "businesswire-generate-series-c-202309" ] }, "insitro": { "id": "insitro", "entityType": "Company", "stage": "Private scaleup", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery", "Human genetics" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Machine-learning drug discovery platform", "In vitro disease models", "Human genetics data platform" ], "scores": { "autonomy": 2, "execution": 3, "verification": 3, "scientificEvidence": 3, "capital": 4, "traction": 3 }, "evidenceLevel": "Commercial partnerships", "accessModel": "Proprietary", "tractionSignals": [ "Daphne Koller founder", "Lilly and Bristol Myers Squibb partnerships", "Reported $2.5B valuation" ], "sourceIds": [ "official-insitro-purpose", "ap-insitro-ai-drug-discovery-202412" ] }, "cellarity": { "id": "cellarity", "entityType": "Company", "stage": "Private scaleup", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery", "Single-cell biology" ], "stackRoles": [ "Reason", "Data", "Verify" ], "products": [ "Cell State-Correcting therapeutics platform", "Single-cell and ML discovery workflows" ], "scores": { "autonomy": 2, "execution": 2, "verification": 3, "scientificEvidence": 2, "capital": 3, "traction": 2 }, "evidenceLevel": "Platform pipeline", "accessModel": "Proprietary", "tractionSignals": [ "$121M Series C", "$274M total disclosed", "Flagship Pioneering origin" ], "sourceIds": [ "official-cellarity-about", "cellarity-series-c-202210" ] }, "abcellera": { "id": "abcellera", "entityType": "Public company", "stage": "Public", "region": "North America", "scienceDomains": [ "Biology", "Antibody discovery", "Clinical manufacturing" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Antibody discovery platform", "Single-cell screening", "Clinical manufacturing facility" ], "scores": { "autonomy": 2, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 4, "traction": 4 }, "evidenceLevel": "Public company", "accessModel": "Partnership platform", "tractionSignals": [ "NASDAQ-listed", "$555.5M IPO proceeds", "DARPA pandemic-response platform", "Eli Lilly COVID antibody collaboration" ], "sourceIds": [ "official-abcellera-platform", "renaissance-abcellera-ipo-202011" ] }, "iambic-therapeutics": { "id": "iambic-therapeutics", "entityType": "Company", "stage": "Clinical-stage platform", "region": "North America", "scienceDomains": [ "Drug discovery", "Chemistry", "Biology" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Enchant", "NeuralPLexer", "AI-driven discovery platform" ], "scores": { "autonomy": 3, "execution": 3, "verification": 3, "scientificEvidence": 3, "capital": 3, "traction": 3 }, "evidenceLevel": "Clinical pipeline", "accessModel": "Proprietary", "tractionSignals": [ "$100M Series B", "NVIDIA collaboration", "Weekly molecular design to biological data loops" ], "sourceIds": [ "official-iambic-platform", "fierce-iambic-series-b-202310" ] }, "causaly": { "id": "causaly", "entityType": "Company", "stage": "Growth", "region": "Europe", "scienceDomains": [ "Biomedical research", "Drug discovery", "Evidence synthesis" ], "stackRoles": [ "Reason", "Verify", "Data" ], "products": [ "Causaly platform", "Agentic Research", "Biomedical knowledge graph" ], "scores": { "autonomy": 3, "execution": 0, "verification": 3, "scientificEvidence": 2, "capital": 3, "traction": 4 }, "evidenceLevel": "Commercial platform", "accessModel": "Proprietary", "tractionSignals": [ "$60M Series B", "Top pharma customers", "500M facts and 70M directional relationships claimed" ], "sourceIds": [ "official-causaly", "finsmes-causaly-series-b-202307" ] }, "iktos": { "id": "iktos", "entityType": "Company", "stage": "Commercial platform", "region": "Europe", "scienceDomains": [ "Chemistry", "Drug discovery", "Medicinal chemistry" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "Makya", "Spaya", "Iktos Robotics Chemistry", "Iktos Robotics Biology" ], "scores": { "autonomy": 3, "execution": 3, "verification": 2, "scientificEvidence": 3, "capital": 2, "traction": 3 }, "evidenceLevel": "Commercial platform", "accessModel": "Hybrid", "tractionSignals": [ "60+ collaborations", "2024 integrated discovery platform launch", "EIC Accelerator support" ], "sourceIds": [ "official-iktos", "official-iktos-about" ] }, "genesis-molecular-ai": { "id": "genesis-molecular-ai", "entityType": "Company", "stage": "Private scaleup", "region": "North America", "scienceDomains": [ "Drug discovery", "Chemistry", "Molecular AI" ], "stackRoles": [ "Reason", "Execute", "Data", "Verify" ], "products": [ "GEMS", "Pearl", "AI-enabled small-molecule pipeline" ], "scores": { "autonomy": 3, "execution": 2, "verification": 3, "scientificEvidence": 3, "capital": 3, "traction": 3 }, "evidenceLevel": "Platform pipeline", "accessModel": "Proprietary", "tractionSignals": [ "$200M Series B", "Lilly and Genentech collaborations", "San Diego integrated lab" ], "sourceIds": [ "official-genesis-molecular-ai", "fierce-genesis-series-b-202308", "cen-genesis-series-b-202308" ] }, "argonne-national-laboratory": { "id": "argonne-national-laboratory", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Materials", "Chemistry", "Biology", "Energy" ], "stackRoles": [ "Reason", "Execute", "Infrastructure", "Data", "Verify" ], "products": [ "Autonomous Discovery initiative", "Polybot", "Self-driving chemist", "Autonomous discovery data platform" ], "scores": { "autonomy": 4, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Research facility", "accessModel": "Public infrastructure", "tractionSignals": [ "Autonomous Discovery program", "Self-driving chemist project", "Polybot materials work" ], "sourceIds": [ "official-argonne-autonomous-discovery", "official-argonne-science101-autonomous-discovery" ] }, "oak-ridge-national-laboratory": { "id": "oak-ridge-national-laboratory", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Materials", "Chemistry", "Energy", "Computing" ], "stackRoles": [ "Reason", "Execute", "Infrastructure", "Data", "Verify" ], "products": [ "Autonomous Science program", "Autonomous Chemistry Lab", "INTERSECT initiative" ], "scores": { "autonomy": 4, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Research facility", "accessModel": "Public infrastructure", "tractionSignals": [ "Autonomous Science page", "Autonomous Chemistry Lab", "Genesis Mission role" ], "sourceIds": [ "official-ornl-autonomous-science", "official-ornl-autonomous-chemistry-lab" ] }, "lawrence-livermore-national-laboratory": { "id": "lawrence-livermore-national-laboratory", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Materials", "Advanced manufacturing", "AI for science" ], "stackRoles": [ "Reason", "Execute", "Infrastructure", "Data", "Verify" ], "products": [ "APEX self-driving alloy lab", "AI/ML research programs", "Exascale computing infrastructure" ], "scores": { "autonomy": 3, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 2 }, "evidenceLevel": "Research facility", "accessModel": "Public infrastructure", "tractionSignals": [ "APEX platform", "Cornell collaboration", "Planned autonomous alloy workflow" ], "sourceIds": [ "official-llnl-apex-202507", "str-llnl-apex-202601" ] }, "nist": { "id": "nist", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Materials", "Metrology", "Formulations", "Biology" ], "stackRoles": [ "Execute", "Verify", "Infrastructure", "Data" ], "products": [ "Autonomous laboratories program", "Autonomous Formulation Lab", "CAMEO", "Hermes" ], "scores": { "autonomy": 3, "execution": 4, "verification": 4, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Research facility", "accessModel": "Public infrastructure", "tractionSignals": [ "Autonomous laboratories portfolio", "NIST SP 1320", "Autonomous Formulation Lab" ], "sourceIds": [ "official-nist-autonomous-labs", "nist-autonomous-systems-materials-202009", "nist-autonomous-formulation-lab-202508" ] }, "nrel": { "id": "nrel", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Materials", "Energy", "Characterization" ], "stackRoles": [ "Execute", "Data", "Verify", "Infrastructure" ], "products": [ "Autonomous experimentation program", "Autonomous synthesis", "Autonomous characterization", "AI-generated control software" ], "scores": { "autonomy": 3, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Research facility", "accessModel": "Public infrastructure", "tractionSignals": [ "Autonomous sputter synthesis", "Autonomous characterization", "Digital Discovery control-software paper" ], "sourceIds": [ "official-nrel-autonomous-experimentation", "rsc-text-to-test-202410" ] }, "brookhaven-national-laboratory": { "id": "brookhaven-national-laboratory", "entityType": "Government lab", "stage": "Research facility", "region": "North America", "scienceDomains": [ "Materials", "Photon science", "Nanoscience" ], "stackRoles": [ "Execute", "Data", "Verify", "Infrastructure" ], "products": [ "NSLS-II autonomous steering tools", "CFN autonomous materials discovery", "Robotics and Bluesky workflows" ], "scores": { "autonomy": 3, "execution": 4, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 3 }, "evidenceLevel": "Research facility", "accessModel": "Public infrastructure", "tractionSignals": [ "NSLS-II DSSI robotics work", "AI-discovered nanostructures", "ILLUMINE autonomous experiment steering" ], "sourceIds": [ "official-bnl-nsls2-dssi", "sciencedaily-bnl-ai-nanostructures-202301" ] }, "materials-genome-initiative": { "id": "materials-genome-initiative", "entityType": "Government initiative", "stage": "Policy and research network", "region": "North America", "scienceDomains": [ "Materials", "Manufacturing", "Policy", "Data infrastructure" ], "stackRoles": [ "Infrastructure", "Data", "Verify" ], "products": [ "Materials Genome Initiative", "Autonomous experimentation reports", "Federal agency coordination" ], "scores": { "autonomy": 1, "execution": 1, "verification": 3, "scientificEvidence": 4, "capital": 2, "traction": 4 }, "evidenceLevel": "Government initiative", "accessModel": "Public infrastructure", "tractionSignals": [ "2011 launch", "Multi-agency federal coordination", "Autonomous experimentation workshop report" ], "sourceIds": [ "official-mgi-about", "official-mgi-autonomous-experimentation-report" ] }, "chai-discovery": { "id": "chai-discovery", "entityType": "Company", "stage": "Growth", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery", "Protein design" ], "stackRoles": [ "Reason", "Data", "Verify" ], "products": [ "Chai-1", "Chai-2", "Molecular design platform" ], "scores": { "autonomy": 3, "execution": 1, "verification": 3, "scientificEvidence": 4, "capital": 4, "traction": 4 }, "evidenceLevel": "Wet-lab validated model", "accessModel": "Commercial partnership", "tractionSignals": [ "$130M Series B at $1.3B valuation", "$225M+ total funding", "Eli Lilly biologics collaboration" ], "sourceIds": [ "official-chai-discovery", "businesswire-chai-series-b-202512", "biospace-chai-lilly-202601", "biorxiv-chai2-zero-shot-202507" ] }, "converge-bio": { "id": "converge-bio", "entityType": "Company", "stage": "Series A", "region": "North America / Middle East", "scienceDomains": [ "Biology", "Drug discovery", "Protein engineering" ], "stackRoles": [ "Reason", "Data", "Verify" ], "products": [ "Target discovery models", "Antibody design system", "Protein manufacturing optimization system" ], "scores": { "autonomy": 2, "execution": 1, "verification": 2, "scientificEvidence": 2, "capital": 3, "traction": 3 }, "evidenceLevel": "Commercial traction", "accessModel": "Commercial partnership", "tractionSignals": [ "$25M Series A", "$30M total raised", "40+ pharma and biotech programs" ], "sourceIds": [ "official-converge-series-a-202601", "techcrunch-converge-series-a-202601" ] }, "generare": { "id": "generare", "entityType": "Company", "stage": "Series A", "region": "Europe", "scienceDomains": [ "Natural products", "Drug discovery", "Synthetic biology" ], "stackRoles": [ "Data", "Execute", "Reason" ], "products": [ "Microbial genome-screening platform", "Cryptic small-molecule dataset", "Synthetic-biology expression workflow" ], "scores": { "autonomy": 2, "execution": 3, "verification": 2, "scientificEvidence": 3, "capital": 3, "traction": 3 }, "evidenceLevel": "Commercial data-generation platform", "accessModel": "Commercial partnership", "tractionSignals": [ "€20M Series A", "200 previously uncharacterized small molecules", "10x dataset scale target by 2027" ], "sourceIds": [ "alven-generare-series-a-202604", "biologydigital-generare-series-a-202604" ] }, "alloy-therapeutics": { "id": "alloy-therapeutics", "entityType": "Company", "stage": "Growth", "region": "North America", "scienceDomains": [ "Biologics", "Drug discovery", "Biotech infrastructure" ], "stackRoles": [ "Execute", "Data", "Infrastructure" ], "products": [ "Biologics discovery platforms", "Integrated preclinical services", "AI and real-world-data infrastructure" ], "scores": { "autonomy": 1, "execution": 3, "verification": 2, "scientificEvidence": 2, "capital": 3, "traction": 4 }, "evidenceLevel": "Commercial infrastructure", "accessModel": "Commercial partnership", "tractionSignals": [ "$40M Series E at $1B valuation", "200+ partners", "22 clinical programs" ], "sourceIds": [ "businesswire-alloy-series-e-202604", "citybiz-alloy-series-e-202604" ] }, "robochem-flex": { "id": "robochem-flex", "entityType": "Academic platform", "stage": "Peer-reviewed prototype", "region": "Europe", "scienceDomains": [ "Chemistry", "Lab automation", "Optimization" ], "stackRoles": [ "Execute", "Reason", "Infrastructure", "Verify" ], "products": [ "RoboChem-Flex modular SDL", "Python control framework", "Bayesian optimization workflow" ], "scores": { "autonomy": 3, "execution": 3, "verification": 3, "scientificEvidence": 4, "capital": 1, "traction": 2 }, "evidenceLevel": "Peer-reviewed prototype", "accessModel": "Open research", "tractionSignals": [ "Nature Synthesis 2026", "Six validation case studies", "Low-cost modular hardware" ], "sourceIds": [ "nature-robochem-flex-202604" ] }, "scivity-labs": { "id": "scivity-labs", "entityType": "Company", "stage": "Pre-seed", "region": "Europe", "scienceDomains": [ "ML research", "Biology", "Chemistry", "General science" ], "stackRoles": [ "Reason", "Execute", "Verify", "Infrastructure" ], "products": [ "Autonomous Laboratory Platform", "Verification surface", "AI for Science Landscape" ], "scores": { "autonomy": 3, "execution": 2, "verification": 2, "scientificEvidence": 1, "capital": 1, "traction": 1 }, "scoreNotes": { "verification": "Conservative score: public materials describe validation outcomes while thresholds, calibration, and check-composition logic remain private." }, "evidenceLevel": "Self-reported platform", "accessModel": "Proprietary", "tractionSignals": [ "Pre-seed company", "Public verification surface", "First autonomous end-to-end research run disclosed" ], "sourceIds": [ "official-scivity-home-202605", "official-scivity-verification-202603", "official-scivity-changelog-202604" ] }, "tetsuwan-scientific": { "id": "tetsuwan-scientific", "entityType": "Company", "stage": "Pre-seed", "region": "North America", "scienceDomains": [ "Biology", "Lab automation", "Cloud lab" ], "stackRoles": [ "Reason", "Execute", "Infrastructure" ], "products": [ "ResearchOS", "Tetsuwan Cloud Lab", "LabAutoWiki" ], "scores": { "autonomy": 2, "execution": 3, "verification": 1, "scientificEvidence": 1, "capital": 1, "traction": 2 }, "evidenceLevel": "Pilot", "accessModel": "Proprietary", "tractionSignals": [ "$2.7M pre-seed", "First lab deployment reported", "ResearchOS execution-layer positioning" ], "sourceIds": [ "official-tetsuwan-home-2026", "techcrunch-tetsuwan-preseed-202412", "synbiobeta-tetsuwan-preseed-202411" ] }, "microsoft-discovery": { "id": "microsoft-discovery", "entityType": "Platform initiative", "stage": "Preview", "region": "North America", "scienceDomains": [ "Biology", "Chemistry", "Materials", "Semiconductors" ], "stackRoles": [ "Reason", "Execute", "Verify", "Data", "Infrastructure" ], "products": [ "Microsoft Discovery Engine", "Azure-governed R&D platform", "Agentic experiment loop" ], "scores": { "autonomy": 3, "execution": 2, "verification": 3, "scientificEvidence": 2, "capital": 4, "traction": 2 }, "evidenceLevel": "Enterprise preview", "accessModel": "Preview / enterprise", "tractionSignals": [ "Build 2025 launch", "Expanded preview in 2026", "Azure-native governance and HPC" ], "sourceIds": [ "official-microsoft-discovery-solution-202604", "official-microsoft-discovery-intro-202505", "official-microsoft-discovery-preview-202604", "techcrunch-microsoft-discovery-202505" ] }, "anthropic": { "id": "anthropic", "entityType": "Frontier AI lab", "stage": "Frontier platform", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery", "General science", "Reasoning" ], "stackRoles": [ "Reason", "Verify", "Data" ], "products": [ "Claude family", "Claude for Life Sciences", "BioMysteryBench", "Agent Skills", "Coefficient Bio team (acquired)" ], "scores": { "autonomy": 3, "execution": 1, "verification": 3, "scientificEvidence": 3, "capital": 4, "traction": 4 }, "evidenceLevel": "Frontier platform", "accessModel": "Proprietary", "tractionSignals": [ "Top of AstaBench scientific-agent leaderboard", "Allen Institute and HHMI founding partnerships", "Coefficient Bio acquisition", "Claude for Life Sciences launch" ], "sourceIds": [ "official-anthropic-claude-life-sciences-202510", "official-anthropic-allen-hhmi-202510", "official-anthropic-biomysterybench-202604", "techcrunch-anthropic-coefficient-bio-202604", "biospace-anthropic-coefficient-202604", "ai2-astabench-update-202604" ] }, "ai2": { "id": "ai2", "entityType": "Nonprofit research institute", "stage": "Research platform", "region": "North America", "scienceDomains": [ "General science", "Literature", "Biology", "Reasoning" ], "stackRoles": [ "Reason", "Verify", "Data" ], "products": [ "Asta agents", "AstaBench", "Scientific Corpus Tool", "OLMo models", "Semantic Scholar" ], "scores": { "autonomy": 3, "execution": 1, "verification": 4, "scientificEvidence": 4, "capital": 3, "traction": 3 }, "evidenceLevel": "Open research", "accessModel": "Open", "tractionSignals": [ "AstaBench ICLR 2026 oral", "UK AISI and General Reasoning adoption", "Submissions from Elicit, SciSpace, Distyl AI, EvoScientist", "Asta runs on 108M+ abstracts and 200M+ paper corpus" ], "sourceIds": [ "official-ai2-asta-launch-202508", "official-ai2-asta-platform", "official-astabench", "ai2-astabench-blog-202508", "ai2-astabench-update-202604", "businesswire-ai2-asta-202508" ] }, "amazon-bio-discovery": { "id": "amazon-bio-discovery", "entityType": "Platform initiative", "stage": "Preview", "region": "North America", "scienceDomains": [ "Biology", "Drug discovery" ], "stackRoles": [ "Reason", "Execute", "Data", "Infrastructure" ], "products": [ "Amazon Bio Discovery", "Biology foundation model catalog", "Lab-in-the-loop orchestration", "Bedrock agents" ], "scores": { "autonomy": 3, "execution": 2, "verification": 2, "scientificEvidence": 2, "capital": 4, "traction": 2 }, "evidenceLevel": "Enterprise preview", "accessModel": "Preview / enterprise", "tractionSignals": [ "April 2026 launch with 40+ BioFMs", "Ginkgo Nebula and Twist Bioscience integration", "Memorial Sloan Kettering early project", "Closed lab-in-the-loop on AWS" ], "sourceIds": [ "official-aws-bio-discovery-blog-202604", "genengnews-aws-bio-discovery-202604", "aipedia-aws-bio-discovery-202604", "official-ginkgo-q1-2026" ] }, "profluent": { "id": "profluent", "entityType": "Company", "stage": "Growth", "region": "North America", "scienceDomains": [ "Biology", "Protein design", "Gene editing" ], "stackRoles": [ "Reason", "Data", "Verify" ], "products": [ "Protein foundation models", "OpenCRISPR-1", "ProGen", "Site-specific recombinase platform" ], "scores": { "autonomy": 2, "execution": 2, "verification": 3, "scientificEvidence": 4, "capital": 4, "traction": 4 }, "evidenceLevel": "Wet-lab validated model", "accessModel": "Hybrid", "tractionSignals": [ "$106M Series B led by Altimeter and Bezos", "$2.25B Lilly recombinase pact", "OpenCRISPR-1 open release", "Nature 2025 AI-designed CRISPR system" ], "sourceIds": [ "official-profluent", "businesswire-profluent-series-b-202511", "biospace-profluent-lilly-202604", "airstreet-profluent-lilly-202604" ] } }, "entity-identifiers": [ { "entity_id": "recursion-pharmaceuticals", "entity_name": "Recursion Pharmaceuticals", "ticker": "RXRX", "cik": "0001601830", "exchange": "Nasdaq", "sec_submissions_url": "https://data.sec.gov/submissions/CIK0001601830.json", "sec_companyfacts_url": "https://data.sec.gov/api/xbrl/companyfacts/CIK0001601830.json", "primary_filing_system": "SEC EDGAR", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-10", "source_ids": [ "sec-rxrx-2026q1" ], "financials_status": "available" }, { "entity_id": "ginkgo-bioworks", "entity_name": "Ginkgo Bioworks", "ticker": "DNA", "cik": "0001830214", "exchange": "NYSE", "sec_submissions_url": "https://data.sec.gov/submissions/CIK0001830214.json", "sec_companyfacts_url": "https://data.sec.gov/api/xbrl/companyfacts/CIK0001830214.json", "primary_filing_system": "SEC EDGAR", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-10", "source_ids": [ "sec-dna-2026q1" ], "financials_status": "available" }, { "entity_id": "abcellera", "entity_name": "AbCellera", "ticker": "ABCL", "cik": "0001703057", "exchange": "Nasdaq", "sec_submissions_url": "https://data.sec.gov/submissions/CIK0001703057.json", "sec_companyfacts_url": "https://data.sec.gov/api/xbrl/companyfacts/CIK0001703057.json", "primary_filing_system": "SEC EDGAR", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-10", "source_ids": [ "sec-abcl-2025-10k" ], "financials_status": "available" }, { "entity_id": "insilico-medicine", "entity_name": "Insilico Medicine", "ticker": "03696.HK", "cik": null, "exchange": "HKEX", "isin": null, "sec_submissions_url": null, "sec_companyfacts_url": null, "primary_filing_system": "HKEX issuer disclosures", "financials_status": "financials_pending", "financials_pending_reason": "Non-SEC public company; HKEX financial statements are not yet normalized into public-financials.json.", "source_tier": "B", "claim_status": "confirmed", "last_verified": "2026-05-10", "source_ids": [ "prnewswire-insilico-hkex-listing-202512" ] }, { "entity_id": "tecan", "entity_name": "Tecan", "ticker": "TECN", "cik": null, "exchange": "SIX Swiss Exchange", "isin": "CH0012100191", "sec_submissions_url": null, "sec_companyfacts_url": null, "primary_filing_system": "SIX issuer reporting", "financials_status": "financials_pending", "financials_pending_reason": "Non-SEC public company; SIX/annual-report fields are not yet normalized into public-financials.json.", "source_tier": "B", "claim_status": "confirmed", "last_verified": "2026-05-10", "source_ids": [ "official-tecan-2025-results-202603" ] } ], "funding": [ { "company_id": "xaira-therapeutics", "date": "2024-04", "company": "Xaira Therapeutics", "amount": "$1B+", "round": "Series A", "lead": "ARCH Venture Partners, Foresite Labs", "notes": "Largest initial biotech commitment in ARCH history. Co-founded by Marc Tessier-Lavigne and Nobel laureate David Baker.", "sourceIds": [ "official-xaira-therapeutics" ] }, { "company_id": "lila-sciences", "date": "2025-03", "company": "Lila Sciences", "amount": "$200M", "round": "Seed", "lead": "Flagship Pioneering", "notes": "Emerged from stealth. General Catalyst, March, ARK, ADIA participated.", "sourceIds": [ "official-lila-sciences" ] }, { "company_id": "isomorphic-labs", "date": "2025-04", "company": "Isomorphic Labs", "amount": "$600M", "round": "First external round", "lead": "Thrive Capital", "notes": "Alphabet spinout focused on AI drug discovery. GV and Alphabet follow-on.", "sourceIds": [ "official-isomorphic-labs" ] }, { "company_id": "lila-sciences", "date": "2025-09", "company": "Lila Sciences", "amount": "$350M", "round": "Series A", "lead": "Braidwell, Collective Global", "notes": "Plus $115M October extension from NVIDIA NVentures, IQT, Analog Devices, Catalio. Total: $550M at $1.3B+ valuation.", "sourceIds": [ "official-lila-sciences" ] }, { "company_id": "periodic-labs", "date": "2025-09", "company": "Periodic Labs", "amount": "~$300M", "round": "Seed", "lead": "Andreessen Horowitz", "notes": "Ex-OpenAI (Liam Fedus) and ex-DeepMind (Ekin Dogus Cubuk) founders. Autonomous materials labs generating new physical-world training data.", "sourceIds": [ "official-periodic-labs" ] }, { "company_id": "chemify", "date": "2025-10", "company": "Chemify", "amount": "$50M+", "round": "Series B", "lead": "Oversubscribed", "notes": "Series B to fund Chemifarm network expansion into Silicon Valley. Universal programmable-synthesis platform from Lee Cronin's Glasgow lab.", "sourceIds": [ "official-chemify" ] }, { "company_id": "sakana-ai", "date": "2025-11", "company": "Sakana AI", "amount": "$135M", "round": "Series B", "lead": "MUFG, Khosla, NEA, Lux, In-Q-Tel, NVIDIA", "notes": "~$2.65B post-money, ~$379M cumulative. Japan's most valuable private startup.", "sourceIds": [ "official-sakana-ai" ] }, { "company_id": "edison-scientific", "date": "2025-12", "company": "Edison Scientific", "amount": "$70M", "round": "Seed", "lead": "Spark Capital, Triatomic", "notes": "FutureHouse commercial spinout. $250M valuation. Commercializes the Kosmos AI co-scientist. Angels: Jeff Dean, Dmitri Alperovitch.", "sourceIds": [ "official-edison-scientific" ] }, { "company_id": "chemlex", "date": "2025-12", "company": "ChemLex", "amount": "$45M", "round": "Growth", "lead": "Granite Asia", "notes": "Raise paired with global HQ move and new self-driving lab in Singapore.", "sourceIds": [ "official-chemlex" ] }, { "company_id": "automata", "date": "2026-01", "company": "Automata", "amount": "$45M", "round": "Series C", "lead": "Dimension", "notes": "Danaher Ventures, Tru Arrow Partners, Octopus Ventures, Entrepreneurs First participated. Lab-automation OS for AI-ready wet labs; five top pharma customers. Danaher joins the board.", "sourceIds": [ "official-automata" ] }, { "company_id": "autoscience-institute", "date": "2026-03", "company": "Autoscience Institute", "amount": "$14M", "round": "Seed", "lead": "General Catalyst", "notes": "Toyota Ventures, Perplexity Fund, S32, MaC participated. Commercial launch of Carl and Mira agents.", "sourceIds": [ "official-autoscience-institute" ] }, { "company_id": "ami-labs", "date": "2026-03", "company": "AMI Labs", "amount": "$1.03B", "round": "Seed", "lead": "Undisclosed syndicate", "notes": "Yann LeCun's new lab. $3.5B pre-money — the largest seed round in European tech history. Signal that frontier-lab veterans are entering the science-AI space outside the Big Tech umbrella.", "sourceIds": [ "official-ami-labs" ] }, { "company_id": "periodic-labs", "date": "2026-03", "company": "Periodic Labs", "amount": "~$7B valuation (talks)", "round": "Secondary / new round", "lead": "Bloomberg report", "notes": "In talks at ~$7B — a 5.4× jump from the $1.3B seed closed six months earlier. Not yet closed; reported by Bloomberg.", "sourceIds": [ "official-periodic-labs" ] }, { "company_id": "helical", "date": "2026-04", "company": "Helical", "amount": "$10M", "round": "Seed", "lead": "redalpine", "notes": "Seed round for a virtual AI lab that turns biological foundation models into reproducible in-silico pharma R&D workflows.", "sourceIds": [ "globenewswire-helical-seed-202604", "techeu-helical-seed-202604" ] }, { "company_id": "10x-science", "date": "2026-04", "company": "10x Science", "amount": "$4.8M", "round": "Seed", "lead": "Initialized Capital", "notes": "Seed round to build AI-native protein characterization for the verification bottleneck in biologic drug development.", "sourceIds": [ "prnewswire-10x-science-seed-202604", "techcrunch-10x-science-seed-202604" ] }, { "company_id": "fathom-therapeutics", "date": "2026-04", "company": "Fathom Therapeutics", "amount": "$47M", "round": "Series A", "lead": "Sutter Hill Ventures", "notes": "Series A for Microcosmos, a physics- and AI-driven small-molecule design platform with lab-in-the-loop discovery programs.", "sourceIds": [ "prnewswire-fathom-series-a-202604", "cen-fathom-series-a-202604" ] }, { "company_id": "kebotix", "date": "2020-04", "company": "Kebotix", "amount": "$11.4M", "round": "Series A", "lead": "Novo Holdings", "notes": "Series A for a self-driving lab platform aimed at rapid chemicals and materials discovery.", "sourceIds": [ "finsmes-kebotix-series-a-202004" ] }, { "company_id": "citrine-informatics", "date": "2023-01", "company": "Citrine Informatics", "amount": "$16M", "round": "Series C", "lead": "Prelude Ventures, Innovation Endeavors", "notes": "Financing to expand Citrine's AI-driven materials and chemical design platform.", "sourceIds": [ "businesswire-citrine-series-c-202301" ] }, { "company_id": "artificial", "date": "2021-05", "company": "Artificial", "amount": "$21.5M", "round": "Series A", "lead": "Microsoft M12", "notes": "Series A for aLab Suite, a lab automation platform for life sciences R&D.", "sourceIds": [ "techcrunch-artificial-series-a-202105" ] }, { "company_id": "synthace", "date": "2021-11", "company": "Synthace", "amount": "$35M", "round": "Series C", "lead": "Horizons Ventures, Sofinnova Partners", "notes": "Series C to expand Synthace's life sciences R&D cloud and experiment-automation platform.", "sourceIds": [ "finsmes-synthace-series-c-202111" ] }, { "company_id": "tetrascience", "date": "2021-04", "company": "TetraScience", "amount": "$80M", "round": "Series B", "lead": "Insight Partners, Alkeon Capital", "notes": "Series B for the Tetra R&D Data Cloud and AI-ready scientific data infrastructure.", "sourceIds": [ "fierce-tetrascience-series-b-202104" ] }, { "company_id": "riffyn", "date": "2019-05", "company": "Riffyn", "amount": "$15M", "round": "Series B", "lead": "M Ventures", "notes": "Series B for Riffyn's scientific process design and R&D data analytics platform.", "sourceIds": [ "biospace-riffyn-series-b-201905" ] }, { "company_id": "generate-biomedicines", "date": "2021-11", "company": "Generate Biomedicines", "amount": "$370M", "round": "Series B", "lead": "Flagship Pioneering and institutional co-investors", "notes": "First external equity raise for Generate's machine-learning-powered generative biology platform.", "sourceIds": [ "fierce-generate-series-b-202111" ] }, { "company_id": "generate-biomedicines", "date": "2023-09", "company": "Generate Biomedicines", "amount": "$273M", "round": "Series C", "lead": "Flagship Pioneering and existing investors", "notes": "Series C to advance a generative AI pipeline of preclinical and clinical protein therapeutics.", "sourceIds": [ "businesswire-generate-series-c-202309" ] }, { "company_id": "cellarity", "date": "2022-10", "company": "Cellarity", "amount": "$121M", "round": "Series C", "lead": "Flagship Pioneering and new strategic investors", "notes": "Series C brought total disclosed funding to $274M for Cellarity's cell-state drug discovery platform.", "sourceIds": [ "cellarity-series-c-202210" ] }, { "company_id": "iambic-therapeutics", "date": "2023-10", "company": "Iambic Therapeutics", "amount": "$100M", "round": "Series B", "lead": "Ascenta Capital, Abingworth", "notes": "Series B to advance AI-discovered therapeutics into clinical development and expand the Iambic platform.", "sourceIds": [ "fierce-iambic-series-b-202310" ] }, { "company_id": "causaly", "date": "2023-07", "company": "Causaly", "amount": "$60M", "round": "Series B", "lead": "ICONIQ Growth", "notes": "Series B for Causaly's AI platform for preclinical discovery and scientific evidence workflows.", "sourceIds": [ "finsmes-causaly-series-b-202307" ] }, { "company_id": "genesis-molecular-ai", "date": "2023-08", "company": "Genesis Molecular AI", "amount": "$200M", "round": "Series B", "lead": "Andreessen Horowitz and unnamed life sciences investor", "notes": "Series B to advance Genesis' AI-enabled drug programs and expand the GEMS platform.", "sourceIds": [ "fierce-genesis-series-b-202308", "cen-genesis-series-b-202308" ] }, { "company_id": "chai-discovery", "date": "2025-12", "company": "Chai Discovery", "amount": "$130M", "round": "Series B", "lead": "Oak HC/FT and General Catalyst", "notes": "Series B valued Chai Discovery at $1.3B, brought total funding above $225M, and followed Chai-2 zero-shot antibody design results.", "sourceIds": [ "businesswire-chai-series-b-202512", "techcrunch-chai-series-b-202512" ] }, { "company_id": "converge-bio", "date": "2026-01", "company": "Converge Bio", "amount": "$25M", "round": "Series A", "lead": "Bessemer Venture Partners", "notes": "Series A brought Converge Bio's total raised to $30M and supported generative-AI systems for target discovery, antibody design, and protein manufacturing optimization.", "sourceIds": [ "official-converge-series-a-202601", "techcrunch-converge-series-a-202601" ] }, { "company_id": "tetsuwan-scientific", "date": "2024-04", "company": "Tetsuwan Scientific", "amount": "$2.7M", "round": "Pre-seed", "lead": "2048 Ventures", "notes": "Oversubscribed pre-seed backing ResearchOS and Tetsuwan's autonomous cloud-lab buildout. Crunchbase lists Apr 18, 2024 as the round date; SynBioBeta and TechCrunch reported the round publicly in late 2024.", "sourceIds": [ "techcrunch-tetsuwan-preseed-202412", "synbiobeta-tetsuwan-preseed-202411", "crunchbase-tetsuwan-financials-202404" ] }, { "company_id": "profluent", "date": "2025-11", "company": "Profluent", "amount": "$106M", "round": "Series B", "lead": "Altimeter Capital and Bezos Expeditions (co-leads)", "notes": "Series B closes total funding at $150M. Spark Capital, Insight Partners, and Air Street Capital participated. Capital scales protein language models and OpenCRISPR-1 adoption.", "sourceIds": [ "businesswire-profluent-series-b-202511" ] } ], "funding-rounds": [ { "entity_id": "xaira-therapeutics", "round_id": "2024-04-xaira-therapeutics-series-a", "entity_name": "Xaira Therapeutics", "close_date": "2024-04", "amount_raw": "$1B+", "amount_currency": "USD", "amount_usd_normalized": 1000000000, "amount_status": "reported_minimum_or_approximate", "round_type": "Series A", "lead_investors": [ "ARCH Venture Partners", "Foresite Labs" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "B", "source_count": 1, "source_ids": [ "official-xaira-therapeutics" ], "included_in_funding_totals": true, "claim_status": "confirmed", "notes": "Largest initial biotech commitment in ARCH history. Co-founded by Marc Tessier-Lavigne and Nobel laureate David Baker.", "last_verified": "2026-05-10" }, { "entity_id": "lila-sciences", "round_id": "2025-03-lila-sciences-seed", "entity_name": "Lila Sciences", "close_date": "2025-03", "amount_raw": "$200M", "amount_currency": "USD", "amount_usd_normalized": 200000000, "amount_status": "reported", "round_type": "Seed", "lead_investors": [ "Flagship Pioneering" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "B", "source_count": 1, "source_ids": [ "official-lila-sciences" ], "included_in_funding_totals": true, "claim_status": "confirmed", "notes": "Emerged from stealth. General Catalyst, March, ARK, ADIA participated.", "last_verified": "2026-05-10" }, { "entity_id": "isomorphic-labs", "round_id": "2025-04-isomorphic-labs-first-external-round", "entity_name": "Isomorphic Labs", "close_date": "2025-04", "amount_raw": "$600M", "amount_currency": "USD", "amount_usd_normalized": 600000000, "amount_status": "reported", "round_type": "First external round", "lead_investors": [ "Thrive Capital" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "B", "source_count": 1, "source_ids": [ "official-isomorphic-labs" ], "included_in_funding_totals": true, "claim_status": "confirmed", "notes": "Alphabet spinout focused on AI drug discovery. GV and Alphabet follow-on.", "last_verified": "2026-05-10" }, { "entity_id": "lila-sciences", "round_id": "2025-09-lila-sciences-series-a", "entity_name": "Lila Sciences", "close_date": "2025-09", "amount_raw": "$350M", "amount_currency": "USD", "amount_usd_normalized": 350000000, "amount_status": "reported", "round_type": "Series A", "lead_investors": [ "Braidwell", "Collective Global" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "B", "source_count": 1, "source_ids": [ "official-lila-sciences" ], "included_in_funding_totals": true, "claim_status": "confirmed", "notes": "Plus $115M October extension from NVIDIA NVentures, IQT, Analog Devices, Catalio. Total: $550M at $1.3B+ valuation.", "last_verified": "2026-05-10" }, { "entity_id": "periodic-labs", "round_id": "2025-09-periodic-labs-seed", "entity_name": "Periodic Labs", "close_date": "2025-09", "amount_raw": "~$300M", "amount_currency": "USD", "amount_usd_normalized": 300000000, "amount_status": "reported_minimum_or_approximate", "round_type": "Seed", "lead_investors": [ "Andreessen Horowitz" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "B", "source_count": 1, "source_ids": [ "official-periodic-labs" ], "included_in_funding_totals": true, "claim_status": "confirmed", "notes": "Ex-OpenAI (Liam Fedus) and ex-DeepMind (Ekin Dogus Cubuk) founders. Autonomous materials labs generating new physical-world training data.", "last_verified": "2026-05-10" }, { "entity_id": "chemify", "round_id": "2025-10-chemify-series-b", "entity_name": "Chemify", "close_date": "2025-10", "amount_raw": "$50M+", "amount_currency": "USD", "amount_usd_normalized": 50000000, "amount_status": "reported_minimum_or_approximate", "round_type": "Series B", "lead_investors": [ "Oversubscribed" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "B", "source_count": 1, "source_ids": [ "official-chemify" ], "included_in_funding_totals": true, "claim_status": "confirmed", "notes": "Series B to fund Chemifarm network expansion into Silicon Valley. Universal programmable-synthesis platform from Lee Cronin's Glasgow lab.", "last_verified": "2026-05-10" }, { "entity_id": "sakana-ai", "round_id": "2025-11-sakana-ai-series-b", "entity_name": "Sakana AI", "close_date": "2025-11", "amount_raw": "$135M", "amount_currency": "USD", "amount_usd_normalized": 135000000, "amount_status": "reported", "round_type": "Series B", "lead_investors": [ "MUFG", "Khosla", "NEA", "Lux", "In-Q-Tel", "NVIDIA" ], "participating_investors": [], "valuation_usd": 2650000000, "valuation_status": "reported", "source_tier": "B", "source_count": 1, "source_ids": [ "official-sakana-ai" ], "included_in_funding_totals": true, "claim_status": "confirmed", "notes": "~$2.65B post-money, ~$379M cumulative. Japan's most valuable private startup.", "last_verified": "2026-05-10" }, { "entity_id": "edison-scientific", "round_id": "2025-12-edison-scientific-seed", "entity_name": "Edison Scientific", "close_date": "2025-12", "amount_raw": "$70M", "amount_currency": "USD", "amount_usd_normalized": 70000000, "amount_status": "reported", "round_type": "Seed", "lead_investors": [ "Spark Capital", "Triatomic" ], "participating_investors": [], "valuation_usd": 250000000, "valuation_status": "reported", "source_tier": "B", "source_count": 1, "source_ids": [ "official-edison-scientific" ], "included_in_funding_totals": true, "claim_status": "confirmed", "notes": "FutureHouse commercial spinout. $250M valuation. Commercializes the Kosmos AI co-scientist. Angels: Jeff Dean, Dmitri Alperovitch.", "last_verified": "2026-05-10" }, { "entity_id": "chemlex", "round_id": "2025-12-chemlex-growth", "entity_name": "ChemLex", "close_date": "2025-12", "amount_raw": "$45M", "amount_currency": "USD", "amount_usd_normalized": 45000000, "amount_status": "reported", "round_type": "Growth", "lead_investors": [ "Granite Asia" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "B", "source_count": 1, "source_ids": [ "official-chemlex" ], "included_in_funding_totals": true, "claim_status": "confirmed", "notes": "Raise paired with global HQ move and new self-driving lab in Singapore.", "last_verified": "2026-05-10" }, { "entity_id": "automata", "round_id": "2026-01-automata-series-c", "entity_name": "Automata", "close_date": "2026-01", "amount_raw": "$45M", "amount_currency": "USD", "amount_usd_normalized": 45000000, "amount_status": "reported", "round_type": "Series C", "lead_investors": [ "Dimension" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "B", "source_count": 1, "source_ids": [ "official-automata" ], "included_in_funding_totals": true, "claim_status": "confirmed", "notes": "Danaher Ventures, Tru Arrow Partners, Octopus Ventures, Entrepreneurs First participated. Lab-automation OS for AI-ready wet labs; five top pharma customers. Danaher joins the board.", "last_verified": "2026-05-10" }, { "entity_id": "autoscience-institute", "round_id": "2026-03-autoscience-institute-seed", "entity_name": "Autoscience Institute", "close_date": "2026-03", "amount_raw": "$14M", "amount_currency": "USD", "amount_usd_normalized": 14000000, "amount_status": "reported", "round_type": "Seed", "lead_investors": [ "General Catalyst" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "B", "source_count": 1, "source_ids": [ "official-autoscience-institute" ], "included_in_funding_totals": true, "claim_status": "confirmed", "notes": "Toyota Ventures, Perplexity Fund, S32, MaC participated. Commercial launch of Carl and Mira agents.", "last_verified": "2026-05-10" }, { "entity_id": "ami-labs", "round_id": "2026-03-ami-labs-seed", "entity_name": "AMI Labs", "close_date": "2026-03", "amount_raw": "$1.03B", "amount_currency": "USD", "amount_usd_normalized": 1030000000, "amount_status": "reported", "round_type": "Seed", "lead_investors": [ "Undisclosed syndicate" ], "participating_investors": [], "valuation_usd": 3500000000, "valuation_status": "reported", "source_tier": "B", "source_count": 1, "source_ids": [ "official-ami-labs" ], "included_in_funding_totals": true, "claim_status": "confirmed", "notes": "Yann LeCun's new lab. $3.5B pre-money — the largest seed round in European tech history. Signal that frontier-lab veterans are entering the science-AI space outside the Big Tech umbrella.", "last_verified": "2026-05-10" }, { "entity_id": "periodic-labs", "round_id": "2026-03-periodic-labs-secondary-new-round", "entity_name": "Periodic Labs", "close_date": "2026-03", "amount_raw": "~$7B valuation (talks)", "amount_currency": "USD", "amount_usd_normalized": null, "amount_status": "undisclosed", "round_type": "Secondary / new round", "lead_investors": [ "Bloomberg report" ], "participating_investors": [], "valuation_usd": 7000000000, "valuation_status": "talks", "source_tier": "C", "source_count": 1, "source_ids": [ "bloomberg-periodic-valuation-talks-202603" ], "included_in_funding_totals": false, "claim_status": "reported", "notes": "In talks at ~$7B — a 5.4× jump from the $1.3B seed closed six months earlier. Not yet closed; reported by Bloomberg.", "last_verified": "2026-05-10" }, { "entity_id": "helical", "round_id": "2026-04-helical-seed", "entity_name": "Helical", "close_date": "2026-04", "amount_raw": "$10M", "amount_currency": "USD", "amount_usd_normalized": 10000000, "amount_status": "reported", "round_type": "Seed", "lead_investors": [ "redalpine" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "C", "source_count": 2, "source_ids": [ "globenewswire-helical-seed-202604", "techeu-helical-seed-202604" ], "included_in_funding_totals": true, "claim_status": "reported", "notes": "Seed round for a virtual AI lab that turns biological foundation models into reproducible in-silico pharma R&D workflows.", "last_verified": "2026-05-10" }, { "entity_id": "10x-science", "round_id": "2026-04-10x-science-seed", "entity_name": "10x Science", "close_date": "2026-04", "amount_raw": "$4.8M", "amount_currency": "USD", "amount_usd_normalized": 4800000, "amount_status": "reported", "round_type": "Seed", "lead_investors": [ "Initialized Capital" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "C", "source_count": 2, "source_ids": [ "prnewswire-10x-science-seed-202604", "techcrunch-10x-science-seed-202604" ], "included_in_funding_totals": true, "claim_status": "reported", "notes": "Seed round to build AI-native protein characterization for the verification bottleneck in biologic drug development.", "last_verified": "2026-05-10" }, { "entity_id": "fathom-therapeutics", "round_id": "2026-04-fathom-therapeutics-series-a", "entity_name": "Fathom Therapeutics", "close_date": "2026-04", "amount_raw": "$47M", "amount_currency": "USD", "amount_usd_normalized": 47000000, "amount_status": "reported", "round_type": "Series A", "lead_investors": [ "Sutter Hill Ventures" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "C", "source_count": 2, "source_ids": [ "prnewswire-fathom-series-a-202604", "cen-fathom-series-a-202604" ], "included_in_funding_totals": true, "claim_status": "reported", "notes": "Series A for Microcosmos, a physics- and AI-driven small-molecule design platform with lab-in-the-loop discovery programs.", "last_verified": "2026-05-10" }, { "entity_id": "kebotix", "round_id": "2020-04-kebotix-series-a", "entity_name": "Kebotix", "close_date": "2020-04", "amount_raw": "$11.4M", "amount_currency": "USD", "amount_usd_normalized": 11400000, "amount_status": "reported", "round_type": "Series A", "lead_investors": [ "Novo Holdings" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "C", "source_count": 1, "source_ids": [ "finsmes-kebotix-series-a-202004" ], "included_in_funding_totals": true, "claim_status": "reported", "notes": "Series A for a self-driving lab platform aimed at rapid chemicals and materials discovery.", "last_verified": "2026-05-10" }, { "entity_id": "citrine-informatics", "round_id": "2023-01-citrine-informatics-series-c", "entity_name": "Citrine Informatics", "close_date": "2023-01", "amount_raw": "$16M", "amount_currency": "USD", "amount_usd_normalized": 16000000, "amount_status": "reported", "round_type": "Series C", "lead_investors": [ "Prelude Ventures", "Innovation Endeavors" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "C", "source_count": 1, "source_ids": [ "businesswire-citrine-series-c-202301" ], "included_in_funding_totals": true, "claim_status": "reported", "notes": "Financing to expand Citrine's AI-driven materials and chemical design platform.", "last_verified": "2026-05-10" }, { "entity_id": "artificial", "round_id": "2021-05-artificial-series-a", "entity_name": "Artificial", "close_date": "2021-05", "amount_raw": "$21.5M", "amount_currency": "USD", "amount_usd_normalized": 21500000, "amount_status": "reported", "round_type": "Series A", "lead_investors": [ "Microsoft M12" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "C", "source_count": 1, "source_ids": [ "techcrunch-artificial-series-a-202105" ], "included_in_funding_totals": true, "claim_status": "reported", "notes": "Series A for aLab Suite, a lab automation platform for life sciences R&D.", "last_verified": "2026-05-10" }, { "entity_id": "synthace", "round_id": "2021-11-synthace-series-c", "entity_name": "Synthace", "close_date": "2021-11", "amount_raw": "$35M", "amount_currency": "USD", "amount_usd_normalized": 35000000, "amount_status": "reported", "round_type": "Series C", "lead_investors": [ "Horizons Ventures", "Sofinnova Partners" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "C", "source_count": 1, "source_ids": [ "finsmes-synthace-series-c-202111" ], "included_in_funding_totals": true, "claim_status": "reported", "notes": "Series C to expand Synthace's life sciences R&D cloud and experiment-automation platform.", "last_verified": "2026-05-10" }, { "entity_id": "tetrascience", "round_id": "2021-04-tetrascience-series-b", "entity_name": "TetraScience", "close_date": "2021-04", "amount_raw": "$80M", "amount_currency": "USD", "amount_usd_normalized": 80000000, "amount_status": "reported", "round_type": "Series B", "lead_investors": [ "Insight Partners", "Alkeon Capital" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "C", "source_count": 1, "source_ids": [ "fierce-tetrascience-series-b-202104" ], "included_in_funding_totals": true, "claim_status": "reported", "notes": "Series B for the Tetra R&D Data Cloud and AI-ready scientific data infrastructure.", "last_verified": "2026-05-10" }, { "entity_id": "riffyn", "round_id": "2019-05-riffyn-series-b", "entity_name": "Riffyn", "close_date": "2019-05", "amount_raw": "$15M", "amount_currency": "USD", "amount_usd_normalized": 15000000, "amount_status": "reported", "round_type": "Series B", "lead_investors": [ "M Ventures" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "C", "source_count": 1, "source_ids": [ "biospace-riffyn-series-b-201905" ], "included_in_funding_totals": true, "claim_status": "reported", "notes": "Series B for Riffyn's scientific process design and R&D data analytics platform.", "last_verified": "2026-05-10" }, { "entity_id": "generate-biomedicines", "round_id": "2021-11-generate-biomedicines-series-b", "entity_name": "Generate Biomedicines", "close_date": "2021-11", "amount_raw": "$370M", "amount_currency": "USD", "amount_usd_normalized": 370000000, "amount_status": "reported", "round_type": "Series B", "lead_investors": [ "Flagship Pioneering and institutional co-investors" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "C", "source_count": 1, "source_ids": [ "fierce-generate-series-b-202111" ], "included_in_funding_totals": true, "claim_status": "reported", "notes": "First external equity raise for Generate's machine-learning-powered generative biology platform.", "last_verified": "2026-05-10" }, { "entity_id": "generate-biomedicines", "round_id": "2023-09-generate-biomedicines-series-c", "entity_name": "Generate Biomedicines", "close_date": "2023-09", "amount_raw": "$273M", "amount_currency": "USD", "amount_usd_normalized": 273000000, "amount_status": "reported", "round_type": "Series C", "lead_investors": [ "Flagship Pioneering and existing investors" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "C", "source_count": 1, "source_ids": [ "businesswire-generate-series-c-202309" ], "included_in_funding_totals": true, "claim_status": "reported", "notes": "Series C to advance a generative AI pipeline of preclinical and clinical protein therapeutics.", "last_verified": "2026-05-10" }, { "entity_id": "cellarity", "round_id": "2022-10-cellarity-series-c", "entity_name": "Cellarity", "close_date": "2022-10", "amount_raw": "$121M", "amount_currency": "USD", "amount_usd_normalized": 121000000, "amount_status": "reported", "round_type": "Series C", "lead_investors": [ "Flagship Pioneering and new strategic investors" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "C", "source_count": 1, "source_ids": [ "cellarity-series-c-202210" ], "included_in_funding_totals": true, "claim_status": "reported", "notes": "Series C brought total disclosed funding to $274M for Cellarity's cell-state drug discovery platform.", "last_verified": "2026-05-10" }, { "entity_id": "iambic-therapeutics", "round_id": "2023-10-iambic-therapeutics-series-b", "entity_name": "Iambic Therapeutics", "close_date": "2023-10", "amount_raw": "$100M", "amount_currency": "USD", "amount_usd_normalized": 100000000, "amount_status": "reported", "round_type": "Series B", "lead_investors": [ "Ascenta Capital", "Abingworth" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "C", "source_count": 1, "source_ids": [ "fierce-iambic-series-b-202310" ], "included_in_funding_totals": true, "claim_status": "reported", "notes": "Series B to advance AI-discovered therapeutics into clinical development and expand the Iambic platform.", "last_verified": "2026-05-10" }, { "entity_id": "causaly", "round_id": "2023-07-causaly-series-b", "entity_name": "Causaly", "close_date": "2023-07", "amount_raw": "$60M", "amount_currency": "USD", "amount_usd_normalized": 60000000, "amount_status": "reported", "round_type": "Series B", "lead_investors": [ "ICONIQ Growth" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "C", "source_count": 1, "source_ids": [ "finsmes-causaly-series-b-202307" ], "included_in_funding_totals": true, "claim_status": "reported", "notes": "Series B for Causaly's AI platform for preclinical discovery and scientific evidence workflows.", "last_verified": "2026-05-10" }, { "entity_id": "genesis-molecular-ai", "round_id": "2023-08-genesis-molecular-ai-series-b", "entity_name": "Genesis Molecular AI", "close_date": "2023-08", "amount_raw": "$200M", "amount_currency": "USD", "amount_usd_normalized": 200000000, "amount_status": "reported", "round_type": "Series B", "lead_investors": [ "Andreessen Horowitz and unnamed life sciences investor" ], "participating_investors": [], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "C", "source_count": 2, "source_ids": [ "fierce-genesis-series-b-202308", "cen-genesis-series-b-202308" ], "included_in_funding_totals": true, "claim_status": "reported", "notes": "Series B to advance Genesis' AI-enabled drug programs and expand the GEMS platform.", "last_verified": "2026-05-10" }, { "entity_id": "chai-discovery", "round_id": "2025-12-chai-discovery-series-b", "entity_name": "Chai Discovery", "close_date": "2025-12", "amount_raw": "$130M", "amount_currency": "USD", "amount_usd_normalized": 130000000, "amount_status": "reported", "round_type": "Series B", "lead_investors": [ "Oak HC/FT", "General Catalyst" ], "participating_investors": [ "Menlo Ventures", "OpenAI", "Dimension", "Thrive Capital", "Emerson Collective", "Glade Brook" ], "valuation_usd": 1300000000, "valuation_status": "reported", "source_tier": "B", "source_count": 2, "source_ids": [ "businesswire-chai-series-b-202512", "techcrunch-chai-series-b-202512" ], "included_in_funding_totals": true, "claim_status": "confirmed", "notes": "Series B valued Chai Discovery at $1.3B and brought total funding to more than $225M.", "last_verified": "2026-05-13" }, { "entity_id": "converge-bio", "round_id": "2026-01-converge-bio-series-a", "entity_name": "Converge Bio", "close_date": "2026-01", "amount_raw": "$25M", "amount_currency": "USD", "amount_usd_normalized": 25000000, "amount_status": "reported", "round_type": "Series A", "lead_investors": [ "Bessemer Venture Partners" ], "participating_investors": [ "TLV Partners", "Vintage Investment Partners", "Saras Capital" ], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "B", "source_count": 2, "source_ids": [ "official-converge-series-a-202601", "techcrunch-converge-series-a-202601" ], "included_in_funding_totals": true, "claim_status": "confirmed", "notes": "Series A brought total funding to $30M and followed commercial traction across target discovery, antibody design, and protein manufacturing optimization.", "last_verified": "2026-05-13" }, { "entity_id": "generare", "round_id": "2026-04-generare-series-a", "entity_name": "Generare", "close_date": "2026-04", "amount_raw": "€20M", "amount_currency": "EUR", "amount_usd_normalized": null, "amount_status": "reported", "round_type": "Series A", "lead_investors": [ "Alven", "Daphni" ], "participating_investors": [ "Galion.exe", "Teampact Ventures", "VIVES Partners" ], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "B", "source_count": 2, "source_ids": [ "alven-generare-series-a-202604", "biologydigital-generare-series-a-202604" ], "included_in_funding_totals": true, "claim_status": "confirmed", "notes": "Series A to scale Generare's proprietary microbial natural-products dataset and drug-discovery platform.", "last_verified": "2026-05-13" }, { "entity_id": "alloy-therapeutics", "round_id": "2026-04-alloy-therapeutics-series-e", "entity_name": "Alloy Therapeutics", "close_date": "2026-04", "amount_raw": "$40M", "amount_currency": "USD", "amount_usd_normalized": 40000000, "amount_status": "reported", "round_type": "Series E", "lead_investors": [ "not disclosed" ], "participating_investors": [ "8VC", "JIC Venture Growth Investments", "Mubadala Capital", "Founders Fund" ], "valuation_usd": 1000000000, "valuation_status": "reported", "source_tier": "B", "source_count": 2, "source_ids": [ "businesswire-alloy-series-e-202604", "citybiz-alloy-series-e-202604" ], "included_in_funding_totals": true, "claim_status": "confirmed", "notes": "Scope-review addition for biotech infrastructure: Alloy reported 200+ partners, 100+ licensed therapeutic programs, and 22 clinical programs.", "last_verified": "2026-05-13" }, { "entity_id": "tetsuwan-scientific", "round_id": "2024-04-tetsuwan-scientific-pre-seed", "entity_name": "Tetsuwan Scientific", "close_date": "2024-04", "amount_raw": "$2.7M", "amount_currency": "USD", "amount_usd_normalized": 2700000, "amount_status": "reported", "round_type": "Pre-seed", "lead_investors": [ "2048 Ventures" ], "participating_investors": [ "Carbon Silicon Ventures", "Referent Ventures", "Everywhere Ventures" ], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "C", "source_count": 3, "source_ids": [ "techcrunch-tetsuwan-preseed-202412", "synbiobeta-tetsuwan-preseed-202411", "crunchbase-tetsuwan-financials-202404" ], "included_in_funding_totals": true, "claim_status": "confirmed", "notes": "Pre-seed financing used to scale Tetsuwan's ResearchOS, cloud-lab roadmap, and first reported rare-disease-lab deployment.", "last_verified": "2026-05-13" }, { "entity_id": "profluent", "round_id": "2025-11-profluent-series-b", "entity_name": "Profluent", "close_date": "2025-11", "amount_raw": "$106M", "amount_currency": "USD", "amount_usd_normalized": 106000000, "amount_status": "reported", "round_type": "Series B", "lead_investors": [ "Altimeter Capital", "Bezos Expeditions" ], "participating_investors": [ "Spark Capital", "Insight Partners", "Air Street Capital" ], "valuation_usd": null, "valuation_status": "undisclosed", "source_tier": "B", "source_count": 1, "source_ids": [ "businesswire-profluent-series-b-202511" ], "included_in_funding_totals": true, "claim_status": "confirmed", "notes": "Series B announcement reports $150M total funding to date and scaling of frontier protein language models and OpenCRISPR-1 adoption.", "last_verified": "2026-05-14" } ], "public-financials": [ { "entity_id": "recursion-pharmaceuticals", "financial_id": "rxrx-2026-q1-sec", "entity_name": "Recursion Pharmaceuticals", "ticker": "RXRX", "cik": "0001601830", "exchange": "Nasdaq", "filing_form": "10-Q", "latest_filing_date": "2026-05-06", "accession_number": "0001601830-26-000078", "report_period_end": "2026-03-31", "period_label": "Q1 2026", "period_type": "quarter", "revenue_usd": 6472000, "r_and_d_expense_usd": 87896000, "cash_and_equivalents_usd": 654473000, "marketable_securities_usd": null, "cash_and_restricted_cash_usd": 665180000, "operating_cash_flow_usd": -81101000, "quarterly_burn_usd": 81101000, "cash_runway_months": 24.2, "debt_usd": 16200000, "debt_basis": "FinanceLeaseLiability; operating lease liabilities excluded from debt_usd.", "collaboration_revenue_usd": null, "segment_notes": "Standardized SEC companyfacts do not isolate collaboration revenue for this period. Company update separately stated runway into early 2028.", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-10", "source_ids": [ "sec-rxrx-2026q1" ], "xbrl_tags": { "revenue": "RevenueFromContractWithCustomerExcludingAssessedTax", "r_and_d_expense": "ResearchAndDevelopmentExpense", "cash_and_equivalents": "CashAndCashEquivalentsAtCarryingValue", "marketable_securities": null, "operating_cash_flow": "NetCashProvidedByUsedInOperatingActivities", "debt": "FinanceLeaseLiability" }, "financials_status": "available" }, { "entity_id": "ginkgo-bioworks", "financial_id": "dna-2026-q1-sec", "entity_name": "Ginkgo Bioworks", "ticker": "DNA", "cik": "0001830214", "exchange": "NYSE", "filing_form": "10-Q", "latest_filing_date": "2026-05-07", "accession_number": "0001628280-26-032116", "report_period_end": "2026-03-31", "period_label": "Q1 2026", "period_type": "quarter", "revenue_usd": 19474000, "r_and_d_expense_usd": 49920000, "cash_and_equivalents_usd": 143864000, "marketable_securities_usd": 229592000, "cash_and_restricted_cash_usd": 188669000, "operating_cash_flow_usd": -46686000, "quarterly_burn_usd": 46686000, "cash_runway_months": 9.2, "debt_usd": null, "debt_basis": "No standardized debt tag in the latest SEC companyfacts accession; operating lease liabilities excluded from debt_usd.", "collaboration_revenue_usd": null, "segment_notes": "Q1 filing follows biosecurity divestiture; cash is split between $143.9M cash and equivalents and $229.6M marketable securities, so liquidity should be read as cash plus securities rather than cash alone.", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-13", "source_ids": [ "official-ginkgo-q1-2026", "sec-dna-2026q1" ], "xbrl_tags": { "revenue": "RevenueFromContractWithCustomerExcludingAssessedTax", "r_and_d_expense": "ResearchAndDevelopmentExpense", "cash_and_equivalents": "CashAndCashEquivalentsAtCarryingValue", "marketable_securities": "MarketableSecuritiesCurrent", "operating_cash_flow": "NetCashProvidedByUsedInOperatingActivities", "debt": null }, "financials_status": "available" }, { "entity_id": "abcellera", "financial_id": "abcl-2025-fy-sec", "entity_name": "AbCellera", "ticker": "ABCL", "cik": "0001703057", "exchange": "Nasdaq", "filing_form": "10-K", "latest_filing_date": "2026-02-24", "accession_number": "0001703057-26-000012", "report_period_end": "2025-12-31", "period_label": "FY 2025", "period_type": "annual", "revenue_usd": 75128000, "r_and_d_expense_usd": 186829000, "cash_and_equivalents_usd": 128513000, "marketable_securities_usd": null, "cash_and_restricted_cash_usd": 153513000, "operating_cash_flow_usd": -131295000, "quarterly_burn_usd": 32823750, "cash_runway_months": 11.7, "debt_usd": null, "debt_basis": "Operating lease liabilities excluded from debt_usd; no standardized debt tag used for this annual record.", "collaboration_revenue_usd": null, "segment_notes": "Latest SEC financial statement available in this pull was an annual 10-K, so quarterly burn is annual operating cash use divided by four.", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-10", "source_ids": [ "sec-abcl-2025-10k" ], "xbrl_tags": { "revenue": "RevenueFromContractWithCustomerExcludingAssessedTax", "r_and_d_expense": "ResearchAndDevelopmentExpense", "cash_and_equivalents": "CashAndCashEquivalentsAtCarryingValue", "marketable_securities": null, "operating_cash_flow": "NetCashProvidedByUsedInOperatingActivities", "debt": null }, "financials_status": "available" } ], "grants-contracts": [ { "entity_id": "ginkgo-bioworks", "award_id": null, "grant_contract_id": "ginkgo-pnnl-m2pc-2025", "source_system": "company_announcement", "agency": "U.S. Department of Energy / Pacific Northwest National Laboratory", "program": "Microbial Molecular Phenotyping Capability (M2PC)", "recipient": "Ginkgo Bioworks", "entity_name": "Ginkgo Bioworks", "amount_obligated_usd": 47000000, "ceiling_amount_usd": 47000000, "period_start": "2025-07", "period_end": null, "topic": "Autonomous microbial molecular phenotyping lab for DOE bioeconomy mission", "source_tier": "B", "claim_status": "reported", "last_verified": "2026-05-10", "source_ids": [ "official-ginkgo-bioworks" ], "notes": "Tracked as non-dilutive contract intelligence, not venture funding. Replace with USAspending or DOE award identifier when available." }, { "entity_id": null, "award_id": "DE-FOA-0003612", "grant_contract_id": "doe-genesis-mission-foa-2026", "source_system": "DOE Funding Opportunity Announcement", "agency": "U.S. Department of Energy", "program": "Genesis Mission National Science and Technology Challenges", "recipient": "Program-level solicitation", "entity_name": null, "amount_obligated_usd": null, "ceiling_amount_usd": 293000000, "period_start": "2026-03", "period_end": null, "topic": "AI-for-science national challenges across manufacturing, biotech, critical materials, nuclear, and quantum", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-10", "source_ids": [ "official-doe-genesis-mission-rfa-202603" ], "notes": "Program-level record retained separately from company funding totals until individual awards and recipients are available." }, { "entity_id": null, "award_id": null, "grant_contract_id": "arpae-catalchem-e-uw-madison-ai-fixcat-202604", "source_system": "ARPA-E project selection", "agency": "U.S. Department of Energy / ARPA-E", "program": "CATALCHEM-E", "recipient": "University of Wisconsin-Madison", "entity_name": null, "amount_obligated_usd": 2835000, "ceiling_amount_usd": 2835000, "period_start": "2026-04", "period_end": null, "topic": "AI-FIXCAT: AI-driven design of industrial fixed-bed reactor catalysts for renewable fuels and chemicals from oxygenated feedstocks", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-13", "source_ids": [ "official-arpae-catalchem-e-projects-202604" ], "notes": "Recipient-level CATALCHEM-E award row; aggregate $34M program event remains in ecosystem-events.json." }, { "entity_id": null, "award_id": null, "grant_contract_id": "arpae-catalchem-e-ames-a-team-202604", "source_system": "ARPA-E project selection", "agency": "U.S. Department of Energy / ARPA-E", "program": "CATALCHEM-E", "recipient": "Ames National Laboratory", "entity_name": null, "amount_obligated_usd": 2520000, "ceiling_amount_usd": 2520000, "period_start": "2026-04", "period_end": null, "topic": "A-TEAM: autonomous high-throughput experimentation to accelerate chemical manufacturing with AI models, robotic synthesis, and multiscale validation", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-13", "source_ids": [ "official-arpae-catalchem-e-projects-202604", "ames-catalchem-e-award-202604" ], "notes": "Recipient-level CATALCHEM-E award row; aggregate $34M program event remains in ecosystem-events.json." }, { "entity_id": null, "award_id": null, "grant_contract_id": "arpae-catalchem-e-ncsu-self-driving-redox-202604", "source_system": "ARPA-E project selection", "agency": "U.S. Department of Energy / ARPA-E", "program": "CATALCHEM-E", "recipient": "North Carolina State University", "entity_name": null, "amount_obligated_usd": 2992500, "ceiling_amount_usd": 2992500, "period_start": "2026-04", "period_end": null, "topic": "Human-AI-robot teaming to accelerate heterogeneous redox catalyst discovery with a self-driving lab", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-13", "source_ids": [ "official-arpae-catalchem-e-projects-202604", "ncsu-catalchem-e-award-202604" ], "notes": "Recipient-level CATALCHEM-E award row; aggregate $34M program event remains in ecosystem-events.json." }, { "entity_id": null, "award_id": null, "grant_contract_id": "arpae-catalchem-e-nlr-synth-on-202604", "source_system": "ARPA-E project selection", "agency": "U.S. Department of Energy / ARPA-E", "program": "CATALCHEM-E", "recipient": "National Laboratory of the Rockies", "entity_name": null, "amount_obligated_usd": 2835000, "ceiling_amount_usd": 2835000, "period_start": "2026-04", "period_end": null, "topic": "SYNTH-ON: autonomous synthetic framework using AI-enabled high-throughput flow synthesis for refinery catalyst discovery", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-13", "source_ids": [ "official-arpae-catalchem-e-projects-202604" ], "notes": "Recipient-level CATALCHEM-E award row; aggregate $34M program event remains in ecosystem-events.json." }, { "entity_id": null, "award_id": null, "grant_contract_id": "arpae-catalchem-e-uconn-react-202604", "source_system": "ARPA-E project selection", "agency": "U.S. Department of Energy / ARPA-E", "program": "CATALCHEM-E", "recipient": "University of Connecticut", "entity_name": null, "amount_obligated_usd": 2916450, "ceiling_amount_usd": 2916450, "period_start": "2026-04", "period_end": null, "topic": "REACT: AI-guided catalyst transformations using design-make-test-learn workflows for methane-to-alcohol electrocatalysts", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-13", "source_ids": [ "official-arpae-catalchem-e-projects-202604" ], "notes": "Recipient-level CATALCHEM-E award row; aggregate $34M program event remains in ecosystem-events.json." }, { "entity_id": null, "award_id": null, "grant_contract_id": "arpae-catalchem-e-pnnl-tap-202604", "source_system": "ARPA-E project selection", "agency": "U.S. Department of Energy / ARPA-E", "program": "CATALCHEM-E", "recipient": "Pacific Northwest National Laboratory", "entity_name": null, "amount_obligated_usd": 2722500, "ceiling_amount_usd": 2722500, "period_start": "2026-04", "period_end": null, "topic": "Trinity AI Platform: automated promotion-gated workflow for inverse catalyst development and industrially relevant CO2-to-ethanol validation", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-13", "source_ids": [ "official-arpae-catalchem-e-projects-202604" ], "notes": "Recipient-level CATALCHEM-E award row; aggregate $34M program event remains in ecosystem-events.json." }, { "entity_id": null, "award_id": null, "grant_contract_id": "arpae-catalchem-e-inl-kinetic-202604", "source_system": "ARPA-E project selection", "agency": "U.S. Department of Energy / ARPA-E", "program": "CATALCHEM-E", "recipient": "Idaho National Laboratory", "entity_name": null, "amount_obligated_usd": 2992500, "ceiling_amount_usd": 2992500, "period_start": "2026-04", "period_end": null, "topic": "KINETIC: robotic high-throughput synthesis and transient testing to learn dynamic catalyst behavior for hydrocarbon conversion", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-13", "source_ids": [ "official-arpae-catalchem-e-projects-202604" ], "notes": "Recipient-level CATALCHEM-E award row; aggregate $34M program event remains in ecosystem-events.json." }, { "entity_id": null, "award_id": null, "grant_contract_id": "arpae-catalchem-e-rochester-catalyst-202604", "source_system": "ARPA-E project selection", "agency": "U.S. Department of Energy / ARPA-E", "program": "CATALCHEM-E", "recipient": "University of Rochester", "entity_name": null, "amount_obligated_usd": 2992500, "ceiling_amount_usd": 2992500, "period_start": "2026-04", "period_end": null, "topic": "CATALYST: language-model-enabled AI workflow for scalable CO2-to-methanol and ethanol catalyst discovery", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-13", "source_ids": [ "official-arpae-catalchem-e-projects-202604" ], "notes": "Recipient-level CATALCHEM-E award row; aggregate $34M program event remains in ecosystem-events.json." }, { "entity_id": "argonne-national-laboratory", "award_id": null, "grant_contract_id": "arpae-catalchem-e-argonne-foundry-202604", "source_system": "ARPA-E project selection", "agency": "U.S. Department of Energy / ARPA-E", "program": "CATALCHEM-E", "recipient": "Argonne National Laboratory", "entity_name": "Argonne National Laboratory", "amount_obligated_usd": 2771100, "ceiling_amount_usd": 2771100, "period_start": "2026-04", "period_end": null, "topic": "Accelerated Catalyst Design Foundry: data library, AI predictions, high-throughput experimentation, and self-driving labs for waste-carbon catalyst design", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-13", "source_ids": [ "official-arpae-catalchem-e-projects-202604" ], "notes": "Recipient-level CATALCHEM-E award row; aggregate $34M program event remains in ecosystem-events.json." }, { "entity_id": null, "award_id": null, "grant_contract_id": "arpae-catalchem-e-lbnl-pandora-202604", "source_system": "ARPA-E project selection", "agency": "U.S. Department of Energy / ARPA-E", "program": "CATALCHEM-E", "recipient": "Lawrence Berkeley National Laboratory", "entity_name": null, "amount_obligated_usd": 2835000, "ceiling_amount_usd": 2835000, "period_start": "2026-04", "period_end": null, "topic": "PANDORA: AI/ML models with high-throughput reactor arrays for atomically tailored CO2-to-commodity-chemical catalysts", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-13", "source_ids": [ "official-arpae-catalchem-e-projects-202604" ], "notes": "Recipient-level CATALCHEM-E award row; aggregate $34M program event remains in ecosystem-events.json." }, { "entity_id": null, "award_id": null, "grant_contract_id": "arpae-catalchem-e-oxylus-sprint-ec-202604", "source_system": "ARPA-E project selection", "agency": "U.S. Department of Energy / ARPA-E", "program": "CATALCHEM-E", "recipient": "Oxylus Energy", "entity_name": null, "amount_obligated_usd": 2955391, "ceiling_amount_usd": 2955391, "period_start": "2026-04", "period_end": null, "topic": "SPRINT-EC: rapid inverse-design network for tailored electrochemical catalysts converting CO2 to methanol", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-13", "source_ids": [ "official-arpae-catalchem-e-projects-202604" ], "notes": "Recipient-level CATALCHEM-E award row; aggregate $34M program event remains in ecosystem-events.json." }, { "entity_id": null, "award_id": null, "grant_contract_id": "arpae-catalchem-e-p2-heat-factory-202604", "source_system": "ARPA-E project selection", "agency": "U.S. Department of Energy / ARPA-E", "program": "CATALCHEM-E", "recipient": "P2 Science, Inc.", "entity_name": null, "amount_obligated_usd": 2834879, "ceiling_amount_usd": 2834879, "period_start": "2026-04", "period_end": null, "topic": "HEAT FACTORY: robotics and machine learning to optimize catalytic transformations of plant-based feedstocks into high-performance liquid fuels", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-13", "source_ids": [ "official-arpae-catalchem-e-projects-202604", "p2science-catalchem-e-award-202604" ], "notes": "Recipient-level CATALCHEM-E award row; aggregate $34M program event remains in ecosystem-events.json." } ], "deals-partnerships": [ { "entity_id": "insilico-medicine", "deal_id": "insilico-lilly-2026-03", "entity_name": "Insilico Medicine", "partner": "Eli Lilly", "date": "2026-03", "deal_type": "AI-driven R&D collaboration", "upfront_cash_usd": 115000000, "potential_milestones_usd": 2750000000, "royalties": "undisclosed royalties reported", "asset_count": null, "therapeutic_area_or_domain": "Preclinical oral therapeutics and Lilly-selected R&D programs", "exclusivity": "worldwide rights to selected portfolio assets", "status": "active", "realized_vs_potential": "Upfront cash is realized; headline value is potential milestones plus royalties.", "source_tier": "B", "claim_status": "confirmed", "source_ids": [ "official-insilico-lilly-202603", "techtarget-insilico-lilly-202603" ], "last_verified": "2026-05-10" }, { "entity_id": "isomorphic-labs", "deal_id": "isomorphic-lilly-2024-01", "entity_name": "Isomorphic Labs", "partner": "Eli Lilly", "date": "2024-01", "deal_type": "AI small-molecule drug discovery collaboration", "upfront_cash_usd": 45000000, "potential_milestones_usd": 1700000000, "royalties": "undisclosed", "asset_count": null, "therapeutic_area_or_domain": "Small-molecule drug discovery", "exclusivity": null, "status": "active", "realized_vs_potential": "Upfront cash is realized; milestone headline is potential value.", "source_tier": "B", "claim_status": "confirmed", "source_ids": [ "official-isomorphic-labs" ], "last_verified": "2026-05-10" }, { "entity_id": "isomorphic-labs", "deal_id": "isomorphic-novartis-2024-01", "entity_name": "Isomorphic Labs", "partner": "Novartis", "date": "2024-01", "deal_type": "AI small-molecule drug discovery collaboration", "upfront_cash_usd": 37500000, "potential_milestones_usd": 1200000000, "royalties": "undisclosed", "asset_count": null, "therapeutic_area_or_domain": "Small-molecule drug discovery", "exclusivity": null, "status": "active", "realized_vs_potential": "Upfront cash is realized; milestone headline is potential value.", "source_tier": "B", "claim_status": "confirmed", "source_ids": [ "official-isomorphic-labs" ], "last_verified": "2026-05-10" }, { "entity_id": "ginkgo-bioworks", "deal_id": "ginkgo-proqr-2026-04", "entity_name": "Ginkgo Bioworks", "partner": "ProQR Therapeutics", "date": "2026-04", "deal_type": "autonomous-lab access partnership", "upfront_cash_usd": null, "potential_milestones_usd": null, "royalties": null, "asset_count": null, "therapeutic_area_or_domain": "RNA editing discovery and high-throughput data generation", "exclusivity": null, "status": "active", "realized_vs_potential": "Commercial terms undisclosed; separated from venture funding totals.", "source_tier": "B", "claim_status": "confirmed", "source_ids": [ "official-proqr-ginkgo-nebula-202604", "rdworld-proqr-ginkgo-nebula-202604" ], "last_verified": "2026-05-10" }, { "entity_id": "ginkgo-bioworks", "deal_id": "openai-ginkgo-cloud-lab-cfps-2026", "entity_name": "Ginkgo Bioworks", "partner": "OpenAI", "date": "2026-02", "deal_type": "closed-loop model-lab integration", "upfront_cash_usd": null, "potential_milestones_usd": null, "royalties": null, "asset_count": null, "therapeutic_area_or_domain": "Cell-free protein synthesis and autonomous biological experimentation", "exclusivity": null, "status": "active", "realized_vs_potential": "Research collaboration and technical integration; commercial terms undisclosed. Reported output included an optimized reaction mix that Ginkgo began selling.", "source_tier": "B", "claim_status": "confirmed", "source_ids": [ "openai-ginkgo-cfps-autonomous-lab-202602", "ginkgo-openai-cfps-202602", "openai-ginkgo-cfps-preprint-202602", "biorxiv-openai-ginkgo-cfps-202602" ], "last_verified": "2026-05-13" }, { "entity_id": "ginkgo-bioworks", "deal_id": "ginkgo-tower-biosecurity-divestiture-2026-04", "entity_name": "Ginkgo Bioworks", "partner": "Tower Biosecurity / Perimeter Systems", "date": "2026-04", "deal_type": "biosecurity divestiture and strategic refocus transaction", "upfront_cash_usd": null, "potential_milestones_usd": null, "royalties": null, "asset_count": null, "therapeutic_area_or_domain": "Biosecurity-segment operations", "exclusivity": null, "status": "closed", "realized_vs_potential": "Ginkgo contributed substantially all Biosecurity-segment operations and received common stock representing about 20% of the purchaser on a fully diluted basis.", "source_tier": "A", "claim_status": "confirmed", "source_ids": [ "official-ginkgo-q1-2026", "sec-dna-2026q1" ], "last_verified": "2026-05-13" }, { "entity_id": "highres-biosolutions", "deal_id": "highres-tetrascience-integration-2024-07", "entity_name": "HighRes Biosolutions", "partner": "TetraScience", "date": "2024-07", "deal_type": "scientific data pipeline integration", "upfront_cash_usd": null, "potential_milestones_usd": null, "royalties": null, "asset_count": null, "therapeutic_area_or_domain": "Automated scientific data pipelines for Cellario-orchestrated labs", "exclusivity": null, "status": "active", "realized_vs_potential": "Commercial terms undisclosed; tracked as partnership evidence rather than realized financing.", "source_tier": "B", "claim_status": "confirmed", "source_ids": [ "official-highres-tetrascience-integration" ], "last_verified": "2026-05-11" }, { "entity_id": "iambic-therapeutics", "deal_id": "iambic-nvidia-platform-2023-10", "entity_name": "Iambic Therapeutics", "partner": "NVIDIA", "date": "2023-10", "deal_type": "AI drug discovery platform collaboration", "upfront_cash_usd": null, "potential_milestones_usd": null, "royalties": null, "asset_count": null, "therapeutic_area_or_domain": "Protein-ligand modeling and accelerated drug discovery platform", "exclusivity": null, "status": "active", "realized_vs_potential": "Commercial terms undisclosed; tracked as partnership evidence rather than realized financing.", "source_tier": "B", "claim_status": "confirmed", "source_ids": [ "fierce-iambic-series-b-202310", "official-iambic-platform" ], "last_verified": "2026-05-11" }, { "entity_id": "abcellera", "deal_id": "abcellera-lilly-covid-2020-03", "entity_name": "AbCellera", "partner": "Eli Lilly", "date": "2020-03", "deal_type": "antibody discovery and development collaboration", "upfront_cash_usd": null, "potential_milestones_usd": null, "royalties": null, "asset_count": null, "therapeutic_area_or_domain": "COVID-19 neutralizing antibody discovery and development", "exclusivity": null, "status": "completed", "realized_vs_potential": "Collaboration produced LY-CoV555/bamlanivimab clinical program; commercial terms were not normalized here.", "source_tier": "B", "claim_status": "confirmed", "source_ids": [ "official-abcellera-lilly-covid-202003", "official-abcellera-platform" ], "last_verified": "2026-05-11" }, { "entity_id": "cellarity", "deal_id": "cellarity-novo-mash-2024-01", "entity_name": "Cellarity", "partner": "Novo Nordisk", "date": "2024-01", "deal_type": "cell-state drug discovery collaboration", "upfront_cash_usd": null, "potential_milestones_usd": null, "royalties": null, "asset_count": null, "therapeutic_area_or_domain": "MASH therapies discovered through cell-state biology", "exclusivity": null, "status": "active", "realized_vs_potential": "Commercial terms undisclosed; tracked as partnership evidence rather than realized financing.", "source_tier": "B", "claim_status": "confirmed", "source_ids": [ "official-cellarity-novo-mash-202401" ], "last_verified": "2026-05-11" }, { "entity_id": "generate-biomedicines", "deal_id": "generate-amgen-multitarget-2022-01", "entity_name": "Generate Biomedicines", "partner": "Amgen", "date": "2022-01", "deal_type": "multi-target protein therapeutics collaboration", "upfront_cash_usd": 50000000, "potential_milestones_usd": 1900000000, "royalties": "future royalties reported; low double-digit royalties reported per program terms", "asset_count": 5, "therapeutic_area_or_domain": "Five initial protein therapeutic programs across multiple modalities, with option for additional programs", "exclusivity": null, "status": "active", "realized_vs_potential": "Upfront cash is realized; $1.9B headline is potential transaction value plus future royalties.", "source_tier": "B", "claim_status": "confirmed", "source_ids": [ "official-generate-amgen-202201" ], "last_verified": "2026-05-11" }, { "entity_id": "insitro", "deal_id": "insitro-bms-fibrosis-2020-11", "entity_name": "Insitro", "partner": "Bristol Myers Squibb", "date": "2020-10", "deal_type": "machine-learning drug discovery collaboration", "upfront_cash_usd": 50000000, "potential_milestones_usd": 2100000000, "royalties": "royalty payments on net product sales reported", "asset_count": null, "therapeutic_area_or_domain": "ALS and frontotemporal dementia target and drug discovery", "exclusivity": null, "status": "active", "realized_vs_potential": "Upfront cash is realized; over $2.1B headline is potential discovery, development, regulatory, and commercial milestones plus royalties.", "source_tier": "B", "claim_status": "confirmed", "source_ids": [ "official-insitro-bms-202010", "official-insitro-bms-extension-202510" ], "last_verified": "2026-05-11" }, { "entity_id": "tetrascience", "deal_id": "tetrascience-highres-integration-2024-07", "entity_name": "TetraScience", "partner": "HighRes Biosolutions", "date": "2024-07", "deal_type": "scientific data pipeline integration", "upfront_cash_usd": null, "potential_milestones_usd": null, "royalties": null, "asset_count": null, "therapeutic_area_or_domain": "Automation-to-data-cloud integration for life-science labs", "exclusivity": null, "status": "active", "realized_vs_potential": "Commercial terms undisclosed; tracked as partnership evidence rather than realized financing.", "source_tier": "B", "claim_status": "confirmed", "source_ids": [ "official-highres-tetrascience-integration", "official-tetrascience-tetra-data" ], "last_verified": "2026-05-11" }, { "entity_id": "automata", "deal_id": "automata-molecular-devices-linq-2026-01", "entity_name": "Automata", "partner": "Molecular Devices", "date": "2026-01", "deal_type": "AI-ready lab automation integration", "upfront_cash_usd": null, "potential_milestones_usd": null, "royalties": null, "asset_count": null, "therapeutic_area_or_domain": "Molecular Devices imaging and detection systems integrated into Automata LINQ for connected lab workflows", "exclusivity": null, "status": "active", "realized_vs_potential": "Commercial terms undisclosed; tracked as instrumentation-integration evidence for Automata's execution layer.", "source_tier": "B", "claim_status": "confirmed", "source_ids": [ "official-molecular-devices-automata-linq-202601", "official-automata-linq" ], "last_verified": "2026-05-13" }, { "entity_id": "chai-discovery", "deal_id": "chai-lilly-biologics-discovery-2026-01", "entity_name": "Chai Discovery", "partner": "Eli Lilly", "date": "2026-01", "deal_type": "AI biologics discovery collaboration", "upfront_cash_usd": null, "potential_milestones_usd": null, "royalties": null, "asset_count": null, "therapeutic_area_or_domain": "Biologics discovery using Chai's frontier AI platform and a custom model trained on proprietary Lilly data", "exclusivity": null, "status": "active", "realized_vs_potential": "Commercial terms undisclosed; tracked as collaboration evidence rather than realized financing.", "source_tier": "B", "claim_status": "confirmed", "source_ids": [ "biospace-chai-lilly-202601", "businesswire-chai-series-b-202512" ], "last_verified": "2026-05-13" }, { "entity_id": "edison-scientific", "deal_id": "edison-nvidia-ai-science-2026-03", "entity_name": "Edison Scientific", "partner": "NVIDIA", "date": "2026-03", "deal_type": "AI scientist infrastructure and benchmark partnership", "upfront_cash_usd": null, "potential_milestones_usd": null, "royalties": null, "asset_count": null, "therapeutic_area_or_domain": "Kosmos training, multimodal literature understanding, and hypothesis-driven bioinformatics benchmarks", "exclusivity": null, "status": "active", "realized_vs_potential": "Commercial terms undisclosed; tracked as a scaling and infrastructure partnership rather than realized financing.", "source_tier": "B", "claim_status": "confirmed", "source_ids": [ "official-edison-nvidia-202603", "nvidia-edison-nemotron-case-study-202603" ], "last_verified": "2026-05-13" }, { "entity_id": "profluent", "deal_id": "profluent-lilly-recombinases-2026-05", "entity_name": "Profluent", "partner": "Eli Lilly", "date": "2026-05", "deal_type": "AI gene-editing strategic research collaboration", "upfront_cash_usd": null, "potential_milestones_usd": 2250000000, "royalties": "tiered royalties on net sales", "asset_count": null, "therapeutic_area_or_domain": "AI-designed site-specific recombinases for diseases with severe unmet need", "exclusivity": null, "status": "active", "realized_vs_potential": "Upfront cash and committed R&D funding undisclosed; up to $2.25B is milestone-contingent across development and commercial milestones plus tiered royalties.", "source_tier": "C", "claim_status": "reported", "source_ids": [ "biospace-profluent-lilly-202604", "airstreet-profluent-lilly-202604" ], "last_verified": "2026-05-14" } ], "ecosystemEvents": [ { "date": "2025-12", "title": "Meta / Manus: Acquisition", "eventType": "Acquisition", "counterparties": [ "Meta", "Manus" ], "amount": "~$2–3B", "amountUsd": null, "lead": "Meta", "summary": "Meta acquired Butterfly Effect (Manus) around Dec 29–30, 2025. Meta to discontinue China operations.", "sourceIds": [ "scivity-curation" ], "counterparty_ids": [] }, { "date": "2026-01", "title": "NVIDIA × Eli Lilly: Partnership", "eventType": "Partnership", "counterparties": [ "NVIDIA", "Eli Lilly" ], "amount": "$1B", "amountUsd": 1000000000, "lead": "NVIDIA, Lilly", "summary": "Co-innovation lab partnership. Picks-and-shovels strategy across pharma.", "sourceIds": [ "official-nvidia-lilly-lab-202601" ], "counterparty_ids": [] }, { "date": "2026-03", "title": "Insilico × Eli Lilly: Deal", "eventType": "Partnership", "counterparties": [ "Insilico Medicine", "Eli Lilly" ], "amount": "$115M upfront ($2.75B deal value)", "amountUsd": 115000000, "lead": "Lilly", "summary": "Expanded AI-driven R&D collaboration gives Lilly worldwide rights to a portfolio of Insilico preclinical oral therapeutics and adds multiple Lilly-selected R&D programs. Terms include $115M upfront plus milestone payments and royalties; only the upfront is treated as committed capital, the $2.75B headline is biobucks.", "sourceIds": [ "official-insilico-lilly-202603", "techtarget-insilico-lilly-202603" ], "counterparty_ids": [ "insilico-medicine" ] }, { "date": "2026-04", "title": "DOE ARPA-E CATALCHEM-E: Government grants", "eventType": "Government program", "counterparties": [ "DOE ARPA-E CATALCHEM-E" ], "amount": "$34M", "amountUsd": 34000000, "lead": "US DOE ARPA-E", "summary": "12 projects pairing AI with self-driving labs to compress catalyst-development timelines from ~10 years to ~1 year. Recipients include NC State ($2.99M), UW-Madison ($2.84M), Ames National Lab ($2.52M).", "sourceIds": [ "official-arpae-catalchem-e-projects-202604", "ncsu-catalchem-e-award-202604" ], "counterparty_ids": [] }, { "date": "2026-04", "title": "Anthropic / Coefficient Bio: Acquisition", "eventType": "Acquisition", "counterparties": [ "Anthropic", "Coefficient Bio" ], "amount": "~$400M", "amountUsd": 400000000, "lead": "Anthropic", "summary": "All-stock. First Anthropic M&A into a new domain. ~10 employees, ~$40M+ per employee. Stanton + Frey (ex-Prescient Design). Dimension VC reportedly 38,513% IRR.", "sourceIds": [ "techcrunch-anthropic-coefficient-bio-202604", "biospace-anthropic-coefficient-202604" ], "counterparty_ids": [ "anthropic" ] }, { "date": "2026-04", "title": "ProQR / Ginkgo Bioworks: Partnership", "eventType": "Partnership", "counterparties": [ "ProQR Therapeutics", "Ginkgo Bioworks" ], "amount": "undisclosed", "amountUsd": null, "lead": "ProQR Therapeutics, Ginkgo Bioworks", "summary": "ProQR gained access to Ginkgo's Nebula autonomous lab for high-throughput data generation supporting Axiomer RNA-editing discovery. Ginkgo also made a strategic equity investment in ProQR.", "sourceIds": [ "official-proqr-ginkgo-nebula-202604", "rdworld-proqr-ginkgo-nebula-202604" ], "counterparty_ids": [ "ginkgo-bioworks" ] }, { "date": "2026-04", "title": "ARIA Activation Partners: Government program", "eventType": "Government program", "counterparties": [ "ARIA" ], "amount": "100M GBP (~$125M) open call", "amountUsd": null, "lead": "ARIA", "summary": "ARIA opened a 100M GBP call for 8-10 Activation Partners, explicitly adding AI-for-science tools, AI Scientist systems, and autonomous labs as capabilities to bring into ARIA-funded R&D. Open call ceiling, not yet committed awards — excluded from capital totals.", "sourceIds": [ "official-aria-activation-partners-202604" ], "counterparty_ids": [] }, { "date": "2026-03", "title": "DOE Genesis Mission: Government grants (FOA)", "eventType": "Government program", "counterparties": [ "DOE Genesis Mission" ], "amount": "$293M FOA ceiling", "amountUsd": null, "lead": "US Department of Energy", "summary": "DE-FOA-0003612. Phase I ($500K–$750K, 9 months) and Phase II ($6M–$15M, 3 years) awards across 20+ national challenges in manufacturing, biotech, critical materials, nuclear, and quantum. Operationalizes the Genesis Mission announced in late 2025. Program ceiling, not yet allocated grants — excluded from capital totals.", "sourceIds": [ "official-doe-genesis-mission-rfa-202603" ], "counterparty_ids": [] }, { "date": "2026-02", "title": "SLAS 2026: lab-automation ecosystem cluster", "eventType": "Conference", "counterparties": [ "SLAS", "Cenevo", "Automata", "ABB Robotics", "Biosero" ], "amount": "n/a", "amountUsd": null, "lead": "SLAS", "summary": "SLAS 2026 in Boston, February 7-11, concentrated lab-automation launches across orchestration software, robotics workcells, and connected AI-ready lab platforms.", "sourceIds": [ "official-slas-2026" ], "counterparty_ids": [ "cenevo", "automata" ] }, { "date": "2026-02", "title": "ABB Robotics SLAS 2026 GoFa workcell demos", "eventType": "Technology demo", "counterparties": [ "ABB Robotics", "Agilent Technologies", "Mettler Toledo" ], "amount": "n/a", "amountUsd": null, "lead": "ABB Robotics", "summary": "ABB showcased three GoFa collaborative robot workcells at SLAS 2026, signaling interoperability with Agilent and Mettler Toledo laboratory instrumentation.", "sourceIds": [ "official-abb-slas-2026" ], "counterparty_ids": [] }, { "date": "2026-02", "title": "Biosero launches GoSimple standardized automation workcells", "eventType": "Product launch", "counterparties": [ "Biosero", "Sartorius", "BD" ], "amount": "n/a", "amountUsd": null, "lead": "Biosero", "summary": "Biosero positioned GoSimple as standardized, pre-validated automation workcells powered by Green Button Go Scheduler, with launch configurations around Sartorius and BD instruments.", "sourceIds": [ "official-biosero-gosimple-202602", "official-biosero-green-button-go-scheduler" ], "counterparty_ids": [] }, { "date": "2026-04", "title": "Project Prometheus: physical-AI capital signal", "eventType": "Adjacent capital signal", "counterparties": [ "Project Prometheus", "JPMorgan", "BlackRock" ], "amount": "$10B (adjacent, physical-AI)", "amountUsd": null, "lead": "Project Prometheus", "summary": "Bloomberg-cited reporting said Project Prometheus closed a roughly $10B round at about a $38B valuation for a physical-AI lab focused on engineering and manufacturing physical products. Tracked as an adjacent ecosystem signal for context, not autonomous-science capital — excluded from capital totals.", "sourceIds": [ "ft-project-prometheus-202604", "investing-project-prometheus-202604" ], "counterparty_ids": [] } ], "milestones": [ { "company_id": "isomorphic-labs", "date": "2021-11", "title": "Isomorphic Labs launches as DeepMind spinout", "summary": "Alphabet announced Isomorphic Labs as a new DeepMind spinout dedicated to AI-first drug discovery. Demis Hassabis serves as CEO while continuing to lead DeepMind; the company builds directly on AlphaFold.", "company": "Isomorphic Labs", "featured": false, "sourceIds": [ "official-isomorphic-labs" ] }, { "company_id": "chemlex", "date": "2022-01", "title": "ChemLex founded as MegaRobo spinout", "summary": "ChemLex was incubated out of Chinese lab-robotics company MegaRobo as a next-generation chemistry CRO — automated synthesis as a service for drug development customers.", "company": "ChemLex", "featured": false, "sourceIds": [ "official-chemlex" ] }, { "company_id": "insilico-medicine", "date": "2022-06", "title": "First end-to-end AI drug enters human trials", "summary": "Insilico dosed the first subject in a Phase 1 trial of ISM001-055 (later named rentosertib) — the first drug whose target was AI-discovered and whose molecule was AI-designed end-to-end to reach human testing.", "company": "Insilico Medicine", "featured": false, "sourceIds": [ "official-insilico-medicine" ] }, { "company_id": "automata", "date": "2023-02", "title": "Automata opens first US office in Boston", "summary": "London-based Automata opened its first North American office in Boston to serve the US life sciences automation market.", "company": "Automata", "featured": false, "sourceIds": [ "official-automata" ] }, { "company_id": "automata", "date": "2023-02", "title": "Automata launches LINQ lab automation platform", "summary": "LINQ debuted at SLAS San Diego 2023 — an open integrated automation platform demonstrating genomics QC and cell counting workflows on modular benches.", "company": "Automata", "featured": false, "sourceIds": [ "official-automata" ] }, { "company_id": "acceleration-consortium", "date": "2023-04", "title": "Acceleration Consortium awarded $200M CFREF grant", "summary": "The University of Toronto received a $200M Canada First Research Excellence Fund grant to support the Acceleration Consortium's self-driving labs — the largest federal research award ever given to a Canadian university.", "company": "Acceleration Consortium", "featured": false, "sourceIds": [ "official-acceleration-consortium" ] }, { "company_id": "strateos", "date": "2023-04", "title": "Strateos pivots to on-site cloud lab deployments", "summary": "Strateos announced a strategic shift to design-and-deploy engagements, letting customers run Strateos-powered automated cloud labs on premises rather than centrally.", "company": "Strateos", "featured": false, "sourceIds": [ "official-strateos" ] }, { "company_id": "automata", "date": "2023-05", "title": "Automata launches LINQ Cloud software", "summary": "Automata released LINQ Cloud, the software layer connecting its automated benches with digital workflows — enabling remote orchestration across distributed life sciences labs.", "company": "Automata", "featured": false, "sourceIds": [ "official-automata" ] }, { "company_id": "recursion-pharmaceuticals", "date": "2023-07", "title": "Recursion + NVIDIA BioNeMo foundation model partnership", "summary": "Recursion and NVIDIA announced a multi-year partnership to train biology and chemistry foundation models on DGX Cloud, with models delivered via BioNeMo — anchored by a $50M NVIDIA PIPE investment.", "company": "Recursion Pharmaceuticals", "featured": false, "sourceIds": [ "official-recursion-pharmaceuticals" ] }, { "company_id": "ginkgo-bioworks", "date": "2023-08", "title": "Ginkgo + Google Cloud five-year AI biology partnership", "summary": "Ginkgo and Google Cloud announced a five-year strategic partnership to build biological foundation models on Vertex AI, spanning genomics, protein function, and synthetic biology.", "company": "Ginkgo Bioworks", "featured": false, "sourceIds": [ "official-ginkgo-bioworks" ] }, { "company_id": "google-deepmind", "date": "2023-09", "title": "DeepMind releases AlphaMissense", "summary": "Published in Science, AlphaMissense classified 89% of 71M possible human missense variants as likely pathogenic or benign, released freely via Ensembl VEP.", "company": "Google DeepMind", "featured": false, "sourceIds": [ "official-google-deepmind" ] }, { "company_id": "a-lab-lbnl", "date": "2023-11", "title": "A-Lab Berkeley publishes in Nature", "summary": "Claimed 41 novel inorganic compounds synthesized autonomously in 17 days. Jointly released with DeepMind's GNoME paper.", "company": "A-Lab (LBNL)", "featured": false, "sourceIds": [ "official-a-lab-lbnl" ] }, { "company_id": "google-deepmind", "date": "2023-11", "title": "DeepMind publishes GNoME: 380k stable crystals", "summary": "The GNoME paper in Nature predicted 2.2M new crystals, including ~380k thermodynamically stable ones. The companion A-Lab paper synthesized a subset autonomously.", "company": "Google DeepMind", "featured": false, "sourceIds": [ "official-google-deepmind" ] }, { "company_id": "recursion-pharmaceuticals", "date": "2023-11", "title": "Recursion + Bayer expand precision oncology alliance", "summary": "Bayer and Recursion refocused their collaboration on precision oncology, covering up to seven programs with potential milestones of $1.5B plus royalties.", "company": "Recursion Pharmaceuticals", "featured": false, "sourceIds": [ "official-recursion-pharmaceuticals" ] }, { "company_id": "emerald-cloud-lab", "date": "2023-12", "title": "Coscientist: GPT-4 autonomous chemistry lab on ECL", "summary": "CMU's Gomes group published Coscientist in Nature — a GPT-4 system that autonomously designed and executed reactions on Emerald Cloud Lab, including palladium cross-couplings. First cloud-lab validation of LLM-driven chemistry.", "company": "Emerald Cloud Lab", "featured": false, "sourceIds": [ "official-emerald-cloud-lab" ] }, { "company_id": "strateos", "date": "2023-12", "title": "Multiply Labs acquires Strateos", "summary": "Robotic biomanufacturing company Multiply Labs acquired Strateos, integrating its automated experimentation platform with cell therapy manufacturing robotics.", "company": "Strateos", "featured": false, "sourceIds": [ "official-strateos" ] }, { "company_id": "pnnl", "date": "2023-12", "title": "PNNL launches Center for AI", "summary": "PNNL established the Center for AI as a virtual research hub coordinating hundreds of scientists across AI for science, security, and energy resilience — led by chief AI scientist Court Corley.", "company": "PNNL", "featured": false, "sourceIds": [ "official-pnnl" ] }, { "company_id": "acceleration-consortium", "date": "2023-12", "title": "Acceleration Consortium + Merck KGaA open-source BayBE", "summary": "The Acceleration Consortium and Merck KGaA released BayBE (Bayesian Back End), an open-source AI experimentation planner under Apache 2.0 — core infrastructure for closed-loop self-driving labs.", "company": "Acceleration Consortium", "featured": false, "sourceIds": [ "official-acceleration-consortium" ] }, { "company_id": "google-deepmind", "date": "2023-12", "title": "DeepMind FunSearch solves open math problem", "summary": "Published in Nature, FunSearch became the first LLM-based system to discover new solutions to the cap set problem — a long-standing open question in extremal combinatorics.", "company": "Google DeepMind", "featured": false, "sourceIds": [ "official-google-deepmind" ] }, { "company_id": "a-lab-lbnl", "date": "2024-01", "title": "Palgrave & Schoop critique A-Lab", "summary": "ChemRxiv critique (arXiv 2402.10359) shows many compounds were already in ICSD and that the system mislabelled disordered phases as novel. Begins the field-wide reckoning with verification.", "company": "A-Lab (LBNL)", "featured": false, "sourceIds": [ "official-a-lab-lbnl" ] }, { "company_id": "google-deepmind", "date": "2024-01", "title": "Isomorphic Labs signs $3B deals with Lilly and Novartis", "summary": "DeepMind spinout Isomorphic Labs announced drug-discovery collaborations with Eli Lilly ($45M upfront, up to $1.7B) and Novartis ($37.5M upfront, up to $1.2B) for AI-designed small molecules.", "company": "Google DeepMind", "featured": false, "sourceIds": [ "official-google-deepmind" ] }, { "company_id": "isomorphic-labs", "date": "2024-01", "title": "Isomorphic signs ~$3B partnerships with Lilly and Novartis", "summary": "Isomorphic announced simultaneous strategic collaborations with Eli Lilly ($45M upfront, up to $1.7B milestones) and Novartis ($37.5M upfront, up to $1.2B milestones) — its first major pharma deals, worth nearly $3B combined.", "company": "Isomorphic Labs", "featured": false, "sourceIds": [ "official-isomorphic-labs" ] }, { "company_id": "atinary", "date": "2024-02", "title": "Atinary strategic partnership with Takeda", "summary": "Atinary announced a collaboration with Takeda to apply its self-driving labs platform to drug discovery; follow-on work reported reaction-yield improvements from <50% to >90% across three closed loops.", "company": "Atinary", "featured": false, "sourceIds": [ "official-atinary" ] }, { "company_id": "sakana-ai", "date": "2024-03", "title": "Sakana publishes Evolutionary Model Merge", "summary": "Sakana used evolutionary algorithms to combine open-source LLMs, producing a Japanese-math model that surpassed larger baselines. Later accepted to Nature Machine Intelligence.", "company": "Sakana AI", "featured": false, "sourceIds": [ "official-sakana-ai" ] }, { "company_id": "a-lab-lbnl", "date": "2024-04", "title": "A-Lab team publishes phase-diagram synthesis strategy", "summary": "Szymanski, Ceder and colleagues published in Nature Synthesis a methodological follow-up reporting 224 robotic reactions across 27 elements — navigating phase diagram complexity to guide robotic inorganic synthesis.", "company": "A-Lab (LBNL)", "featured": false, "sourceIds": [ "official-a-lab-lbnl" ] }, { "company_id": "emerald-cloud-lab", "date": "2024-04", "title": "CMU Cloud Lab opens on Emerald infrastructure", "summary": "The 16,000 sq ft Carnegie Mellon Cloud Lab in Pittsburgh's Bakery Square opened for PI access, powered by Emerald Cloud Lab and housing 130+ instrument types for remote-run chemistry and biology.", "company": "Emerald Cloud Lab", "featured": false, "sourceIds": [ "official-emerald-cloud-lab" ] }, { "company_id": "xaira-therapeutics", "date": "2024-04", "title": "Xaira emerges from stealth with Tessier-Lavigne as CEO", "summary": "Xaira Therapeutics launched publicly with former Stanford president Marc Tessier-Lavigne as CEO and David Baker's lab as a co-founding scientific anchor — the largest initial biotech commitment in ARCH history.", "company": "Xaira Therapeutics", "featured": false, "sourceIds": [ "official-xaira-therapeutics" ] }, { "company_id": "ginkgo-bioworks", "date": "2024-04", "title": "Ginkgo + Novo Nordisk expand alliance across R&D", "summary": "Ginkgo and Novo Nordisk expanded their partnership under a five-year framework covering manufacturing scale-up and early protein discovery for diabetes and obesity pipelines.", "company": "Ginkgo Bioworks", "featured": false, "sourceIds": [ "official-ginkgo-bioworks" ] }, { "company_id": "google-deepmind", "date": "2024-05", "title": "DeepMind releases AlphaFold 3", "summary": "Published in Nature, AlphaFold 3 extended protein structure prediction to DNA, RNA, ligands, and ions. The free AlphaFold Server generated over 1M structures within six months.", "company": "Google DeepMind", "featured": false, "sourceIds": [ "official-google-deepmind" ] }, { "company_id": "acceleration-consortium", "date": "2024-05", "title": "Acceleration Consortium distributed SDL paper in Science", "summary": "Published in Science, 'Delocalized, Asynchronous, Closed-Loop Discovery of Organic Laser Emitters' demonstrated cloud-orchestrated self-driving labs across Canada, Scotland, USA, and Japan — discovering 21 new organic solid-state laser materials.", "company": "Acceleration Consortium", "featured": false, "sourceIds": [ "official-acceleration-consortium" ] }, { "company_id": "recursion-pharmaceuticals", "date": "2024-05", "title": "Recursion completes BioHive-2 supercomputer", "summary": "Recursion completed BioHive-2, an NVIDIA DGX SuperPOD with 504 H100 GPUs delivering ~2 exaflops. Ranked #35 on TOP500 and the fastest supercomputer owned by any pharma company.", "company": "Recursion Pharmaceuticals", "featured": false, "sourceIds": [ "official-recursion-pharmaceuticals" ] }, { "company_id": "ginkgo-bioworks", "date": "2024-06", "title": "Ginkgo announces 35% workforce reduction", "summary": "Ginkgo began cutting at least 35% of its staff (~400 employees) targeting $200M in operating expense reduction and facility consolidation, aiming for EBITDA breakeven by 2026.", "company": "Ginkgo Bioworks", "featured": false, "sourceIds": [ "official-ginkgo-bioworks" ] }, { "company_id": "google-deepmind", "date": "2024-07", "title": "AlphaProof + AlphaGeometry 2 hit IMO silver medal level", "summary": "DeepMind's AlphaProof (formal reasoning) and AlphaGeometry 2 jointly solved 4 of 6 problems at the 2024 International Mathematical Olympiad, matching silver-medalist performance.", "company": "Google DeepMind", "featured": false, "sourceIds": [ "official-google-deepmind" ] }, { "company_id": "openai", "date": "2024-07", "title": "OpenAI + Los Alamos bioscience lab partnership", "summary": "OpenAI and Los Alamos National Laboratory announced a partnership to evaluate GPT-4o in physical bioscience lab settings — the first multimodal biosecurity evaluation of its kind.", "company": "OpenAI", "featured": false, "sourceIds": [ "official-openai" ] }, { "company_id": "atinary", "date": "2024-07", "title": "Atinary + ACD/Labs AI informatics partnership", "summary": "Atinary partnered with ACD/Labs to integrate self-driving labs with ACD/Labs' chemistry informatics software for planning and analysis of high-throughput and parallel chemistry experiments.", "company": "Atinary", "featured": false, "sourceIds": [ "official-atinary" ] }, { "company_id": "sakana-ai", "date": "2024-08", "title": "Sakana publishes 'The AI Scientist'", "summary": "First credible end-to-end demonstration of AI generating ML research papers — $15 per paper. Raises the question: can autonomous research clear peer review?", "company": "Sakana AI", "featured": false, "sourceIds": [ "official-sakana-ai" ] }, { "company_id": "acceleration-consortium", "date": "2024-08", "title": "Accelerate 2024 conference at UBC", "summary": "The third annual Accelerate conference drew 350+ researchers from academia, industry, and government to the University of British Columbia — the Acceleration Consortium's flagship SDL community gathering.", "company": "Acceleration Consortium", "featured": false, "sourceIds": [ "official-acceleration-consortium" ] }, { "company_id": "google-deepmind", "date": "2024-09", "title": "DeepMind unveils AlphaProteo", "summary": "DeepMind's first AI system to design novel high-affinity protein binders for targets including VEGF-A and the SARS-CoV-2 spike RBD — with 3x-300x better binding than prior methods.", "company": "Google DeepMind", "featured": false, "sourceIds": [ "official-google-deepmind" ] }, { "company_id": "recursion-pharmaceuticals", "date": "2024-09", "title": "Recursion SYCAMORE Phase 2 mixed CCM readout", "summary": "Phase 2 data for REC-994 in cerebral cavernous malformation met the primary safety endpoint but showed only modest efficacy signals — a watched readout for Recursion's lead AI-designed drug.", "company": "Recursion Pharmaceuticals", "featured": false, "sourceIds": [ "official-recursion-pharmaceuticals" ] }, { "company_id": "emerald-cloud-lab", "date": "2024-09", "title": "Emerald Cloud Lab opens 105,000 sq ft Austin facility", "summary": "ECL moved from South San Francisco to a 105,000 sq ft Austin lab — seven times larger — holding $30M+ of equipment and ~230 automated instruments capable of 500 concurrent experiments.", "company": "Emerald Cloud Lab", "featured": false, "sourceIds": [ "official-emerald-cloud-lab" ] }, { "company_id": "ginkgo-bioworks", "date": "2024-09", "title": "Ginkgo launches AA-0 protein LLM and Model API", "summary": "Ginkgo released its first protein LLM (ginkgo-AA-0-650m), trained on 2B+ proprietary protein sequences, alongside a Model API for programmatic access on Google Cloud.", "company": "Ginkgo Bioworks", "featured": false, "sourceIds": [ "official-ginkgo-bioworks" ] }, { "company_id": "openai", "date": "2024-09", "title": "OpenAI o1 exceeds PhD baseline on GPQA", "summary": "OpenAI released o1, its first reasoning model, scoring ~78% on GPQA-Diamond physics/chemistry/biology questions — surpassing the ~70% PhD-expert baseline.", "company": "OpenAI", "featured": false, "sourceIds": [ "official-openai" ] }, { "company_id": "futurehouse", "date": "2024-09", "title": "FutureHouse publishes PaperQA2", "summary": "FutureHouse released PaperQA2, the first AI agent to exceed PhD- and postdoc-level biology researchers on literature retrieval tasks (LitQA2). Open-sourced on GitHub.", "company": "FutureHouse", "featured": false, "sourceIds": [ "official-futurehouse" ] }, { "company_id": "google-deepmind", "date": "2024-10", "title": "Hassabis and Jumper win Nobel Prize in Chemistry", "summary": "Demis Hassabis and John Jumper shared the 2024 Nobel Prize in Chemistry (jointly with David Baker) for AlphaFold's protein structure prediction — the first Nobel to recognize an AI-driven scientific breakthrough.", "company": "Google DeepMind", "featured": false, "sourceIds": [ "official-google-deepmind" ] }, { "company_id": "google-deepmind", "date": "2024-11", "title": "DeepMind publishes AlphaQubit in Nature", "summary": "A transformer-based decoder for surface-code quantum errors, AlphaQubit outperformed prior state-of-the-art decoders on Sycamore hardware data — joint work with Google Quantum AI.", "company": "Google DeepMind", "featured": false, "sourceIds": [ "official-google-deepmind" ] }, { "company_id": "recursion-pharmaceuticals", "date": "2024-11", "title": "Recursion closes Exscientia merger", "summary": "Recursion completed its ~$650M all-stock acquisition of Exscientia, combining biology-at-scale imaging with precision chemistry — creating the largest end-to-end AI-native drug discovery platform.", "company": "Recursion Pharmaceuticals", "featured": false, "sourceIds": [ "official-recursion-pharmaceuticals" ] }, { "company_id": "ginkgo-bioworks", "date": "2024-11", "title": "Ginkgo hits first milestone in Merck biologics deal", "summary": "Ginkgo achieved its first milestone under the up-to-$490M Merck biologics manufacturing partnership, triggering a $9M research payment and advancing to stage two.", "company": "Ginkgo Bioworks", "featured": false, "sourceIds": [ "official-ginkgo-bioworks" ] }, { "company_id": "acceleration-consortium", "date": "2024-11", "title": "Lash Miller building expansion groundbreaking", "summary": "The University of Toronto broke ground on the Lash Miller Chemical Laboratories expansion — the permanent home for the Acceleration Consortium's self-driving labs, backed by $130M in U of T investment.", "company": "Acceleration Consortium", "featured": false, "sourceIds": [ "official-acceleration-consortium" ] }, { "company_id": "atinary", "date": "2024-11", "title": "Atinary Lab launch with Takeda expansion", "summary": "Atinary unveiled Atinary Lab, a physical self-driving laboratory for AI-driven drug discovery and chemistry — expanding its Takeda collaboration into an in-house facility.", "company": "Atinary", "featured": false, "sourceIds": [ "official-atinary" ] }, { "company_id": "google-deepmind", "date": "2024-12", "title": "DeepMind GenCast beats ENS weather forecasting", "summary": "DeepMind's diffusion-based weather model, published in Nature, produced more accurate 15-day forecasts than ECMWF's ENS 97% of the time, including extreme events like tropical cyclones.", "company": "Google DeepMind", "featured": false, "sourceIds": [ "official-google-deepmind" ] }, { "company_id": "acceleration-consortium", "date": "2024-12", "title": "Acceleration Consortium joins IBM/Meta AI Alliance", "summary": "The Acceleration Consortium joined the AI Alliance — 140+ organizations co-founded by IBM and Meta — to develop open-source AI tools and foundation models for accelerating scientific discovery.", "company": "Acceleration Consortium", "featured": false, "sourceIds": [ "official-acceleration-consortium" ] }, { "company_id": "recursion-pharmaceuticals", "date": "2024-12", "title": "Recursion REC-617 CDK7 Phase 1 interim data", "summary": "Interim ELUCIDATE Phase 1 monotherapy data for REC-617 showed a durable partial response in platinum-resistant ovarian cancer with favorable tolerability; combination studies planned for H1 2025.", "company": "Recursion Pharmaceuticals", "featured": false, "sourceIds": [ "official-recursion-pharmaceuticals" ] }, { "company_id": "openai", "date": "2025-01", "title": "OpenAI + Retro Biosciences: 50x Yamanaka factor gains", "summary": "OpenAI and Retro Biosciences revealed GPT-4b micro, a protein-engineering model producing redesigned Sox2/Klf4 variants with >50-fold higher pluripotency marker expression.", "company": "OpenAI", "featured": false, "sourceIds": [ "official-openai" ] }, { "company_id": "openai", "date": "2025-01", "title": "OpenAI + DOE National Laboratories partnership", "summary": "OpenAI agreed to give up to 15,000 scientists across all 17 DOE National Laboratories access to its o-series reasoning models for disease research, math, physics, and nuclear security.", "company": "OpenAI", "featured": false, "sourceIds": [ "official-openai" ] }, { "company_id": "google-deepmind", "date": "2025-02", "title": "DeepMind publishes AI co-scientist", "summary": "Gemini-based multi-agent system (Generation / Reflection / Ranking / Evolution / Proximity / Meta-review agents) using Elo tournaments. arXiv:2502.18864.", "company": "Google DeepMind", "featured": false, "sourceIds": [ "official-google-deepmind" ] }, { "company_id": "openai", "date": "2025-02", "title": "OpenAI launches Deep Research", "summary": "o3-powered autonomous research agent. Adds MCP and trusted-source restrictions by Feb 2026.", "company": "OpenAI", "featured": false, "sourceIds": [ "official-openai" ] }, { "company_id": "autoscience-institute", "date": "2025-03", "title": "Autoscience Carl accepted at ICLR workshops", "summary": "3 of 4 submissions passed meta-review. First autonomous system with a non-negative-results workshop acceptance.", "company": "Autoscience Institute", "featured": false, "sourceIds": [ "official-autoscience-institute" ] }, { "company_id": "lila-sciences", "date": "2025-03", "title": "Lila Sciences emerges from stealth", "summary": "Flagship Pioneering unveiled Lila Sciences as a scientific superintelligence platform with AI Science Factories spanning life, chemical, and materials sciences. George Church joined as Chief Scientist.", "company": "Lila Sciences", "featured": false, "sourceIds": [ "official-lila-sciences" ] }, { "company_id": "sakana-ai", "date": "2025-03", "title": "Sakana AI Scientist-v2 paper passes ICLR workshop review", "summary": "The first fully AI-generated paper to pass peer review at an ICLR 2025 workshop (scores 6/7/6, avg 6.33). Sakana withdrew it post-acceptance, citing missing disclosure norms.", "company": "Sakana AI", "featured": false, "sourceIds": [ "official-sakana-ai" ] }, { "company_id": "zeon-systems", "date": "2025-04", "title": "Zeon Systems launches natural-language lab automation", "summary": "YC X25 startup Zeon launched publicly with natural-language-to-robot-arm lab automation, with active pilots at UCSF and Stanford for nanoparticle screens and clinical assay pipetting.", "company": "Zeon Systems", "featured": false, "sourceIds": [ "official-zeon-systems" ] }, { "company_id": "radical-ai", "date": "2025-04", "title": "Radical AI open-sources TorchSim", "summary": "Radical AI released TorchSim, a PyTorch-native atomistic simulation engine for the MLIP era, offering up to 100x speedup over ASE with GPU-accelerated batched simulations.", "company": "Radical AI", "featured": false, "sourceIds": [ "official-radical-ai" ] }, { "company_id": "xaira-therapeutics", "date": "2025-04", "title": "Xaira hires scGPT creator Bo Wang", "summary": "Xaira appointed Bo Wang, creator of the scGPT single-cell foundation model, as SVP and Head of Biomedical AI — leading virtual cell and multimodal biology model development.", "company": "Xaira Therapeutics", "featured": false, "sourceIds": [ "official-xaira-therapeutics" ] }, { "company_id": "google-deepmind", "date": "2025-05", "title": "DeepMind unveils AlphaEvolve", "summary": "AlphaEvolve, a Gemini-powered coding agent for algorithm discovery, found a 4x4 complex matrix multiplication using 48 multiplications — beating Strassen's 1969 bound — and improved Google datacenter scheduling by 0.7%.", "company": "Google DeepMind", "featured": false, "sourceIds": [ "official-google-deepmind" ] }, { "company_id": "openai", "date": "2025-05", "title": "OpenAI launches HealthBench", "summary": "A benchmark of 5,000 realistic health conversations graded by rubrics from 262 physicians across 60 countries. o3 scored 60% vs GPT-3.5's 16%.", "company": "OpenAI", "featured": false, "sourceIds": [ "official-openai" ] }, { "company_id": "futurehouse", "date": "2025-05", "title": "FutureHouse launches Crow/Falcon/Owl/Phoenix platform", "summary": "FutureHouse publicly released its platform with four specialized scientific agents accessible via web and API — opening AI research agents to scientists worldwide.", "company": "FutureHouse", "featured": false, "sourceIds": [ "official-futurehouse" ] }, { "company_id": "futurehouse", "date": "2025-05", "title": "FutureHouse Robin proposes ripasudil for dry AMD", "summary": "Robin — a multi-agent system combining Crow, Falcon, and Finch — autonomously generated the hypothesis and experimental design identifying ripasudil as a novel candidate for dry age-related macular degeneration.", "company": "FutureHouse", "featured": false, "sourceIds": [ "official-futurehouse" ] }, { "company_id": "google-deepmind", "date": "2025-06", "title": "DeepMind launches AlphaGenome", "summary": "AlphaGenome, a unified model predicting thousands of regulatory properties from DNA sequences up to 1M bp, launched via API for non-commercial research.", "company": "Google DeepMind", "featured": false, "sourceIds": [ "official-google-deepmind" ] }, { "company_id": "sakana-ai", "date": "2025-06", "title": "Sakana releases AB-MCTS and open-source TreeQuest", "summary": "Adaptive Branching MCTS with open-source TreeQuest enabled frontier models (Gemini 2.5 Pro, o4-mini, DeepSeek-R1) to cooperate at inference, solving 30%+ of problems unsolvable by any single model alone.", "company": "Sakana AI", "featured": false, "sourceIds": [ "official-sakana-ai" ] }, { "company_id": "futurehouse", "date": "2025-06", "title": "FutureHouse releases ether0 chemistry reasoning model", "summary": "ether0, a 24B-parameter open-weights reasoning model for chemistry, outperformed frontier LLMs and human experts on molecular design tasks. Featured on Nature.com.", "company": "FutureHouse", "featured": false, "sourceIds": [ "official-futurehouse" ] }, { "company_id": "xaira-therapeutics", "date": "2025-06", "title": "Xaira releases X-Atlas/Orion Perturb-seq dataset", "summary": "Xaira released an 8M-cell genome-wide Perturb-seq atlas covering ~20,000 protein-coding genes via its FiCS platform — at release, the largest publicly available Perturb-seq dataset for AI training.", "company": "Xaira Therapeutics", "featured": false, "sourceIds": [ "official-xaira-therapeutics" ] }, { "company_id": "chemify", "date": "2025-06", "title": "Chemify opens Chemifarm synthesis facility in Glasgow", "summary": "Chemify launched its first Chemifarm — a £12M, 21,500 sq ft autonomous chemputation facility in Maryhill, Glasgow — described as the world's first commercial chemputation site.", "company": "Chemify", "featured": false, "sourceIds": [ "official-chemify" ] }, { "company_id": "insilico-medicine", "date": "2025-06", "title": "Rentosertib Phase 2a results published in Nature Medicine", "summary": "Insilico published positive Phase 2a results for rentosertib in idiopathic pulmonary fibrosis — the first proof-of-concept clinical validation of an AI-discovered target and AI-designed molecule. 60mg cohort showed +98.4mL FVC change vs -20.3mL placebo.", "company": "Insilico Medicine", "featured": false, "sourceIds": [ "official-insilico-medicine" ] }, { "company_id": "ginkgo-bioworks", "date": "2025-07", "title": "Ginkgo + PNNL $47M deal to build M2PC autonomous lab", "summary": "Ginkgo was selected by PNNL to build the Microbial Molecular Phenotyping Capability (M2PC), a 97-RAC, 32,000 sq ft autonomous lab supporting DOE's bioeconomy mission.", "company": "Ginkgo Bioworks", "featured": false, "sourceIds": [ "official-ginkgo-bioworks" ] }, { "company_id": "xaira-therapeutics", "date": "2025-07", "title": "Xaira appoints Jeff Jonker as President and COO", "summary": "Xaira named Jeff Jonker President and Chief Operating Officer as the company scaled operations following its 2024 launch.", "company": "Xaira Therapeutics", "featured": false, "sourceIds": [ "official-xaira-therapeutics" ] }, { "company_id": "radical-ai", "date": "2025-08", "title": "Radical AI wins AFWERX contract for hypersonic alloys", "summary": "Radical AI was awarded a ~$1.2M AFWERX Direct-to-Phase II contract to accelerate discovery of high-entropy alloys with enhanced thermomechanical properties for hypersonic flight.", "company": "Radical AI", "featured": false, "sourceIds": [ "official-radical-ai" ] }, { "company_id": "lila-sciences", "date": "2025-09", "title": "Lila Sciences signs 235,500 sq ft Cambridge lab lease", "summary": "Lila Sciences signed the largest Greater Boston lab lease of Q3 2025, at Alewife Park in Cambridge, to house its expanding network of AI Science Factories.", "company": "Lila Sciences", "featured": false, "sourceIds": [ "official-lila-sciences" ] }, { "company_id": "anthropic", "date": "2025-10", "title": "Anthropic launches Claude for Life Sciences", "summary": "Benchling, PubMed, 10x Genomics, ClinicalTrials.gov, Synapse.org connectors. Agent Skills for life-sciences workflows. Allen Institute and HHMI named as founding scientific partners.", "company": "Anthropic", "featured": false, "sourceIds": [ "official-anthropic-claude-life-sciences-202510", "official-anthropic-allen-hhmi-202510" ] }, { "company_id": "periodic-labs", "date": "2025-10", "title": "Periodic Labs emerges from stealth", "summary": "Ex-OpenAI VP Liam Fedus and ex-DeepMind materials scientist Ekin Dogus Cubuk unveiled Periodic Labs with a $300M seed to build autonomous labs for superconductors and novel materials — with space, defense, and semiconductor customers named at launch.", "company": "Periodic Labs", "featured": false, "sourceIds": [ "official-periodic-labs" ] }, { "company_id": "openai", "date": "2025-10", "title": "OpenAI for Science team launch", "summary": "OpenAI formally announced an OpenAI for Science initiative led by Kevin Weil, recruiting academics including theoretical physicist Alex Lupsasca to accelerate scientific research using GPT-5.", "company": "OpenAI", "featured": false, "sourceIds": [ "official-openai" ] }, { "company_id": "futurehouse", "date": "2025-11", "title": "FutureHouse spins out Edison Scientific and launches Kosmos", "summary": "FutureHouse spun off commercial entity Edison Scientific to serve pharma clients and simultaneously launched Kosmos, an AI co-scientist the team says runs work equivalent to six months of PhD research in a single pass.", "company": "FutureHouse", "featured": false, "sourceIds": [ "official-futurehouse" ] }, { "company_id": "thesis", "date": "2025-11", "title": "Thesis hits state-of-the-art on OpenAI's MLE-Bench", "summary": "Thesis (YC F25) reached state-of-the-art on OpenAI's MLE-Bench with a 48.44% medal rate (vs the prior ~17% best) — achieved in one month with only $10k of compute.", "company": "Thesis", "featured": false, "sourceIds": [ "official-thesis" ] }, { "company_id": "ami-labs", "date": "2025-11", "title": "Yann LeCun departs Meta for world-model startup", "summary": "Meta's chief AI scientist Yann LeCun announced on November 19 that he was leaving after a decade to found an independent lab focused on world models — a public defection from the LLM consensus at a frontier lab.", "company": "AMI Labs", "featured": false, "sourceIds": [ "official-ami-labs" ] }, { "company_id": "edison-scientific", "date": "2025-11", "title": "Edison Scientific spins out of FutureHouse", "summary": "FutureHouse announced Edison Scientific, its for-profit commercial arm, to scale the AI-scientist stack beyond the nonprofit's research scope. Sam Rodriques is CEO; Andrew White is co-founder.", "company": "Edison Scientific", "featured": false, "sourceIds": [ "official-edison-scientific" ] }, { "company_id": "edison-scientific", "date": "2025-11", "title": "Kosmos AI co-scientist launches with seven demonstrated discoveries", "summary": "Edison released Kosmos (arXiv 2511.02824) — a multi-agent AI scientist that executes ~42,000 lines of code and reads ~1,500 papers per run. Launch showcased seven discoveries across neuroscience, materials, and statistical genetics: three reproducing human-led findings and four entirely novel.", "company": "Edison Scientific", "featured": false, "sourceIds": [ "official-edison-scientific" ] }, { "company_id": "google-deepmind", "date": "2025-12", "title": "DOE × DeepMind Genesis Mission", "summary": "AI co-scientist deployed across all 17 US National Labs via Gemini for Government. Stated goal: double US scientific productivity within a decade.", "company": "Google DeepMind", "featured": false, "sourceIds": [ "official-google-deepmind" ] }, { "company_id": "recursion-pharmaceuticals", "date": "2025-12", "title": "Recursion TUPELO Phase 1b/2 shows 43% polyp reduction in FAP", "summary": "REC-4881 produced a median 43% reduction in polyp burden in familial adenomatous polyposis at 25 weeks, with durable effects off therapy. Recursion planned to discuss a registration pathway with FDA.", "company": "Recursion Pharmaceuticals", "featured": false, "sourceIds": [ "official-recursion-pharmaceuticals" ] }, { "company_id": "pnnl", "date": "2025-12", "title": "DOE Secretary commissions PNNL AMP2 autonomous platform", "summary": "Secretary of Energy Chris Wright commissioned the Anaerobic Microbial Phenotyping Platform (AMP2) at PNNL's EMSL — reportedly the world's largest autonomous-capable system for anaerobic microbial experimentation.", "company": "PNNL", "featured": false, "sourceIds": [ "official-pnnl" ] }, { "company_id": "autoscience-institute", "date": "2025-12", "title": "Autoscience Carl wins Silver in Kaggle Santa 2025", "summary": "Autoscience's autonomous system secured a Silver Medal in Kaggle's Santa 2025 competition against 3,300 teams — the first time a fully autonomous AI system placed in a live featured Kaggle competition.", "company": "Autoscience Institute", "featured": false, "sourceIds": [ "official-autoscience-institute" ] }, { "company_id": "openai", "date": "2025-12", "title": "OpenAI launches FrontierScience benchmark", "summary": "FrontierScience, an expert-written benchmark of 700+ physics/chemistry/biology questions split into Olympiad and Research tracks. GPT-5.2 led with 77% (Olympiad) and 25% (Research).", "company": "OpenAI", "featured": false, "sourceIds": [ "official-openai" ] }, { "company_id": "radical-ai", "date": "2025-12", "title": "Radical AI signs DOE MOU for Genesis Mission", "summary": "Radical AI was among 24 organizations signing MOUs with the U.S. Department of Energy to support the Genesis Mission for AI-accelerated scientific discovery.", "company": "Radical AI", "featured": false, "sourceIds": [ "official-radical-ai" ] }, { "company_id": "insilico-medicine", "date": "2025-12", "title": "Insilico lists on Hong Kong Stock Exchange", "summary": "Insilico Medicine listed on the HKEX on December 30, raising ~$293M at the largest Hong Kong biotech IPO of 2025. The retail tranche was oversubscribed 1,427× — the first AI-driven biotech on the exchange. Cornerstone investors included Lilly, Tencent, and Temasek.", "company": "Insilico Medicine", "featured": false, "sourceIds": [ "official-insilico-medicine" ] }, { "company_id": "chemlex", "date": "2025-12", "title": "ChemLex opens Singapore self-driving lab", "summary": "ChemLex announced its new global headquarters and flagship self-driving chemistry lab in Singapore, alongside a $45M round led by Granite Asia. Signed an MoU with Singapore's Experimental Drug Development Centre (EDDC). Customer base reached 70+, including six of the top ten global pharma companies.", "company": "ChemLex", "featured": false, "sourceIds": [ "official-chemlex" ] }, { "company_id": "radical-ai", "date": "2026-01", "title": "Radical AI opens NY's first autonomous materials lab", "summary": "Radical AI established New York's first fully autonomous materials science laboratory at the Brooklyn Navy Yard, with capacity for ~100 AI-driven experiments per day.", "company": "Radical AI", "featured": false, "sourceIds": [ "official-radical-ai" ] }, { "company_id": "openai", "date": "2026-01", "title": "OpenAI launches Prism workspace for scientists", "summary": "Prism, a free LaTeX-native collaborative research workspace powered by GPT-5.2, targets drafting, revising, and publishing scientific papers.", "company": "OpenAI", "featured": false, "sourceIds": [ "official-openai" ] }, { "company_id": "autoscience-institute", "date": "2026-01", "title": "Autoscience introduces Mira ML research agent", "summary": "Autoscience announced Mira, an automated ML research scientist designed to invent new model architectures. Mira was entered into a live $50k Kaggle competition and climbed to a Bronze Medal autonomously.", "company": "Autoscience Institute", "featured": false, "sourceIds": [ "official-autoscience-institute" ] }, { "company_id": "a-lab-lbnl", "date": "2026-02", "title": "Nature issues A-Lab correction", "summary": "Formal correction (650(8100):E1). Paper no longer claims materials necessarily novel to science. The field's defining verification failure is officially recorded.", "company": "A-Lab (LBNL)", "featured": false, "sourceIds": [ "official-a-lab-lbnl" ] }, { "company_id": "automata", "date": "2026-02", "title": "Beckman Coulter + Automata strategic partnership", "summary": "Danaher subsidiary Beckman Coulter Life Sciences integrated its liquid handling and genomic instruments into Automata's Linq platform, announced at SLAS 2026.", "company": "Automata", "featured": false, "sourceIds": [ "official-automata" ] }, { "company_id": "cenevo", "date": "2025-07", "title": "Labguru and Titian rebrand as Cenevo", "summary": "Labguru and Titian Software rebranded as Cenevo, combining ELN/LIMS, sample management, workflow automation, and informatics into a lab orchestration platform with 950+ customer organizations, 45,000+ users, and 8 of the top 10 pharma companies.", "company": "Cenevo", "featured": false, "sourceIds": [ "official-cenevo-rebrand-202507" ] }, { "company_id": "cenevo", "date": "2026-02", "title": "Cenevo launches AI protocol and automation agents", "summary": "Cenevo introduced AI Protocol Conversion and AI Automation agents for translating lab protocols into executable workflows with human approvals, audit trails, and FDA 21 CFR Part 11 / GxP compliance positioning.", "company": "Cenevo", "featured": false, "sourceIds": [ "rdworld-cenevo-ai-agents-202602" ] }, { "company_id": "ginkgo-bioworks", "date": "2026-02", "title": "GPT-5 drives 36,000 CFPS experiments in Ginkgo Cloud Lab", "summary": "OpenAI and Ginkgo reported a closed-loop autonomous CFPS optimization study on February 5: GPT-5 designed experiments, Ginkgo's Cloud Lab executed them with minimal human intervention, results returned to GPT-5, and the next round was proposed by the model. The campaign covered 36,000+ unique reaction compositions across 580 automated plates, six closed-loop rounds, about 40% lower sfGFP production cost, and a 27% titer increase.", "company": "Ginkgo Bioworks", "featured": true, "sourceIds": [ "openai-ginkgo-cfps-autonomous-lab-202602", "ginkgo-openai-cfps-202602", "openai-ginkgo-cfps-preprint-202602", "biorxiv-openai-ginkgo-cfps-202602" ] }, { "company_id": "ginkgo-bioworks", "date": "2026-02", "title": "Ginkgo pivots to autonomous labs, divests biosecurity", "summary": "In its Q4 2025 report, Ginkgo announced a strategic pivot to make autonomous labs its core offering and plans to divest biosecurity to concentrate capital on the platform.", "company": "Ginkgo Bioworks", "featured": false, "sourceIds": [ "official-ginkgo-bioworks" ] }, { "company_id": "google-deepmind", "date": "2026-02", "title": "Isomorphic Labs unveils IsoDDE drug-design engine", "summary": "Isomorphic Labs announced IsoDDE, a unified drug-design platform reported to more than double AlphaFold 3's accuracy on protein-ligand structure prediction benchmarks.", "company": "Google DeepMind", "featured": false, "sourceIds": [ "official-google-deepmind" ] }, { "company_id": "isomorphic-labs", "date": "2026-02", "title": "Isomorphic unveils IsoDDE drug design engine", "summary": "Isomorphic announced IsoDDE — an integrated drug design engine claimed to more than double AlphaFold 3 on protein-ligand binding prediction and to cover the full discovery workflow end-to-end. Pipeline of 17 programs now spans oncology, immunology, and cardiovascular disease.", "company": "Isomorphic Labs", "featured": false, "sourceIds": [ "official-isomorphic-labs" ] }, { "company_id": "sakana-ai", "date": "2026-03", "title": "Sakana explainer paper in Nature", "summary": "Nature 651:914–919. A human-authored description of the AI Scientist system, commonly misreported as an AI-authored Nature paper.", "company": "Sakana AI", "featured": false, "sourceIds": [ "official-sakana-ai" ] }, { "company_id": "pnnl", "date": "2026-03", "title": "PNNL unveils Autonomy Studio", "summary": "PNNL publicly detailed the Autonomy Studio — a purpose-built LDRD-funded facility for design-build-test-learn cycles with digital twins, AI agents, and robotic platforms across biological engineering and critical minerals.", "company": "PNNL", "featured": false, "sourceIds": [ "official-pnnl" ] }, { "company_id": "xaira-therapeutics", "date": "2026-03", "title": "Xaira launches X-Cell virtual cell model", "summary": "Xaira released X-Cell, a 4.9B-parameter diffusion-based virtual cell model trained on X-Atlas/Pisces — the largest-ever CRISPRi Perturb-seq dataset at 25.6M single-cell transcriptomes.", "company": "Xaira Therapeutics", "featured": false, "sourceIds": [ "official-xaira-therapeutics" ] }, { "company_id": "ami-labs", "date": "2026-03", "title": "AMI Labs emerges from stealth with JEPA-first world-model bet", "summary": "Yann LeCun's Paris-based lab launched publicly with a $1.03B seed at $3.5B pre-money — the largest seed in European tech history. Architecture bet is explicitly against the LLM paradigm, building on Joint Embedding Predictive Architecture (JEPA). Research bench includes Saining Xie, Pascale Fung, and Michael Rabbat.", "company": "AMI Labs", "featured": false, "sourceIds": [ "official-ami-labs" ] }, { "company_id": "anthropic", "date": "2026-04", "title": "Anthropic acquires Coefficient Bio", "summary": "All-stock, ~$400M. First Anthropic M&A into a new domain. Signals that frontier labs will enter AI-for-science domain-by-domain through acqui-hires.", "company": "Anthropic", "featured": false, "sourceIds": [ "techcrunch-anthropic-coefficient-bio-202604", "biospace-anthropic-coefficient-202604" ] }, { "company_id": "openai", "date": "2026-04", "title": "OpenAI publishes Industrial Policy paper", "summary": "Calls for 'distributed AI-enabled laboratories' as public infrastructure. Commits $100K grants + $1M API credits, not capital to build the network.", "company": "OpenAI", "featured": false, "sourceIds": [ "official-openai" ] }, { "company_id": null, "date": "2026-04", "title": "DOE ARPA-E CATALCHEM-E awards $34M to 12 AI + self-driving lab projects", "summary": "First large federal bet explicitly tying autonomous labs to industrial catalyst development. Targets compressing catalyst discovery from ~10 years to ~1 year across universities and national labs.", "company": null, "featured": true, "sourceIds": [ "scivity-curation" ] }, { "company_id": "sakana-ai", "date": "2026-04", "title": "Sakana AI launches Marlin closed beta", "summary": "First commercial product from Sakana. Autonomous enterprise research agent (Adaptive Branching MCTS + AI Scientist workflow) producing slide decks and multi-dozen-page reports in one 8-hour run — the AI Scientist architecture repurposed for corporate strategy.", "company": "Sakana AI", "featured": false, "sourceIds": [ "official-sakana-ai" ] }, { "company_id": "chemify", "date": "2026-04", "title": "Chemify publishes three chemputation validation papers", "summary": "Cronin/Chemify published three peer-reviewed papers in PNAS and Nature Communications sister journals validating the chemputer for autonomous drug synthesis and kinase inhibitor discovery.", "company": "Chemify", "featured": false, "sourceIds": [ "official-chemify" ] }, { "company_id": "atinary", "date": "2026-02", "title": "Atinary opens Boston self-driving lab", "summary": "Atinary announced its Boston AI-powered lab with two autonomous Scientific Discovery Factories running closed-loop Design-Make-Test-Analyze-Learn cycles. The lab integrates Atinary's no-code AI platform with robotics and instruments from ABB, Agilent, Bruker, Chemspeed, and Mettler-Toledo for small-molecule synthesis and catalysis.", "company": "Atinary", "featured": true, "sourceIds": [ "official-atinary-boston-sdl-202602", "selectscience-atinary-boston-sdl-202602" ] }, { "company_id": "ginkgo-bioworks", "date": "2026-03", "title": "Ginkgo launches Ginkgo Cloud Lab on Nebula", "summary": "Ginkgo opened web-browser access to Nebula, its 50+ instrument autonomous lab in Boston, via Ginkgo Cloud Lab. Protocols are submitted in natural language and assessed for fleet compatibility by an AI agent called EstiMate. Ginkgo plans to move all R&D services onto Nebula and decommission traditional benches.", "company": "Ginkgo Bioworks", "featured": false, "sourceIds": [ "official-ginkgo-bioworks" ] }, { "company_id": null, "date": "2026-03", "title": "DOE opens $293M Genesis Mission funding solicitation", "summary": "DOE published FOA DE-FOA-0003612 for the Genesis Mission, inviting Phase I and Phase II proposals across 20+ AI-for-science national challenges. Phase I awards up to $750K for nine-month projects; Phase II up to $15M over three years. Applications due April 28, 2026.", "company": null, "featured": false, "sourceIds": [ "official-doe-genesis-mission-rfa-202603" ] }, { "company_id": null, "date": "2026-03", "title": "Nature publishes \"Inside the self-driving lab revolution\"", "summary": "A Nature Technology Feature (30 March 2026) surveys the maturing self-driving-lab industry — profiling King's Eve at Chalmers, CMU's Coscientist on Emerald Cloud Lab, and the commercial wave behind them — while acknowledging the limits that still require human scientists.", "company": null, "featured": false, "sourceIds": [ "scivity-curation" ] }, { "company_id": null, "date": "2026-04", "title": "Stanford AI Index 2026: AI agents trail PhD scientists on complex tasks", "summary": "Stanford HAI's 2026 AI Index Report (released April 13) found that the best frontier AI agents score roughly half as well as human PhDs on complex multi-step scientific workflows, even as models passed the PhD-expert baseline on GPQA. A Nature commentary framed it as the field's first quantitative reality check on autonomous-science agents.", "company": null, "featured": false, "sourceIds": [ "scivity-curation" ] }, { "company_id": "futurehouse", "date": "2026-04", "title": "FutureHouse introduces DISCO enzyme design model", "summary": "FutureHouse, Caltech, Mila, and collaborators introduced DISCO, a multimodal generative model that co-designs protein sequence and structure. The team reported functional new-to-nature carbene-transfer enzymes, including a B-H insertion design reaching 98% product and 5,170 total turnovers.", "company": "FutureHouse", "featured": false, "sourceIds": [ "official-futurehouse-disco-202604", "arxiv-disco-enzyme-design-202604" ] }, { "company_id": "sakana-ai", "date": "2026-04", "title": "Sakana AI opens Fugu multi-agent beta", "summary": "Sakana AI introduced Fugu, a multi-agent orchestration system that learns to route work across frontier foundation models and itself. Sakana frames the beta as an API product for coding, mathematics, and scientific-reasoning workloads.", "company": "Sakana AI", "featured": false, "sourceIds": [ "official-sakana-fugu-202604" ] }, { "company_id": null, "date": "2026-04", "title": "Agent4Science launches as agents-only research social network", "summary": "University of Chicago's Chenhao Tan launched Agent4Science, a Reddit-style site where AI agents share, debate, and review papers — including AI-generated ones — while humans can only observe. An early experiment in agent-to-agent scientific discourse described in Nature on April 20.", "company": null, "featured": false, "sourceIds": [ "scivity-curation" ] }, { "company_id": "isomorphic-labs", "date": "2026-04", "title": "Isomorphic Labs publicly preparing first-in-human AI-designed drug trials", "summary": "Speaking at WIRED Health on April 16, Isomorphic Labs president Max Jaderberg confirmed the AlphaFold-derived oncology pipeline is gearing up to enter the clinic, the first major operational signal that an AI-native drug-discovery program is reaching first-in-human studies.", "company": "Isomorphic Labs", "featured": false, "sourceIds": [ "creati-isomorphic-trials-202604", "clinicaltrialsarena-isomorphic-trials-202604" ] }, { "company_id": "openai", "date": "2026-04", "title": "OpenAI launches GPT-Rosalind for life sciences research", "summary": "OpenAI introduced GPT-Rosalind on April 16, a frontier reasoning model tuned for biology, drug discovery, and translational medicine, alongside a Codex life-sciences plugin connecting to 50+ scientific tools. Available as a research preview through OpenAI's trusted access program with Amgen, Moderna, Thermo Fisher, and the Allen Institute as launch partners.", "company": "OpenAI", "featured": true, "sourceIds": [ "openai-gpt-rosalind-202604", "fiercebiotech-gpt-rosalind-202604", "axios-gpt-rosalind-202604" ] }, { "company_id": "google-deepmind", "date": "2026-04", "title": "Google DeepMind launches Deep Research and Deep Research Max", "summary": "Google DeepMind released Deep Research and Deep Research Max in public preview on April 21, autonomous research agents built on Gemini 3.1 Pro that combine open-web search with private data via Model Context Protocol. Deep Research Max runs roughly 160 search queries per task, targeting long-horizon research workflows.", "company": "Google DeepMind", "featured": false, "sourceIds": [ "blog-google-deep-research-max-202604", "edtechinnovationhub-deep-research-max-202604" ] }, { "company_id": "openai", "date": "2026-04", "title": "OpenAI releases GeneBench scientific-agent benchmark", "summary": "OpenAI released GeneBench on April 23, a benchmark of 103 multi-stage scientific data analysis tasks across 10 genomics and quantitative biology domains. Frontier models score around 25-33%, with OpenAI noting that individual tasks often correspond to multi-day projects for expert computational biologists.", "company": "OpenAI", "featured": false, "sourceIds": [ "openai-genebench-202604", "biorxiv-genebench-202604" ] }, { "company_id": "recursion-pharmaceuticals", "date": "2026-05", "title": "Recursion Q1 2026: REC-4881 establishes clinical proof of concept in FAP", "summary": "Recursion's Q1 2026 results, reported May 6, established clinical proof of concept for REC-4881 in familial adenomatous polyposis with significant precancerous polyp reduction; Recursion has initiated FDA engagement for a potential registrational study. The company also booked a fifth Sanofi milestone payment on a partnered first-in-class oncology program and reaffirmed runway into early 2028.", "company": "Recursion Pharmaceuticals", "featured": false, "sourceIds": [ "official-recursion-q1-2026", "fool-recursion-q1-2026" ] }, { "company_id": "ginkgo-bioworks", "date": "2026-05", "title": "Ginkgo positions Nebula as its autonomous-lab core", "summary": "Ginkgo's Q1 2026 results, reported May 7, highlighted Cloud Lab adoption from ProQR and Amazon's Bio Discovery platform. Jason Kelly positioned Nebula as the world's largest autonomous lab and targeted doubling the Nebula footprint in 2026. Q1 revenue was $19M with $373M in cash, cash equivalents and marketable securities.", "company": "Ginkgo Bioworks", "featured": false, "sourceIds": [ "official-ginkgo-q1-2026", "stocktitan-ginkgo-q1-2026" ] }, { "company_id": "ginkgo-bioworks", "date": "2026-04", "title": "Ginkgo closes Biosecurity divestiture to Tower Biosecurity", "summary": "On April 3, 2026, Ginkgo completed the divestiture of substantially all Biosecurity-segment operations to Tower Biosecurity, also known as Perimeter Systems, receiving common stock representing about 20% of the purchaser on a fully diluted basis and refocusing the company around autonomous labs.", "company": "Ginkgo Bioworks", "featured": false, "sourceIds": [ "official-ginkgo-q1-2026", "sec-dna-2026q1" ] }, { "company_id": "edison-scientific", "date": "2026-05", "title": "Edison Scientific publishes LABBench2 biology research benchmark", "summary": "Edison Scientific released LABBench2 on May 2, a 1,892-task benchmark spanning literature, databases, sequences, protocols, patents, trials, and source quality across 11 task families. Moves agent evaluation from short-form biology recall toward practical research workflows including retrieval, file handling, and tool use.", "company": "Edison Scientific", "featured": false, "sourceIds": [ "official-edison-labbench2-202605", "huggingface-labbench2-202603" ] }, { "company_id": "coscientist-cmu", "date": "2023-12", "title": "Coscientist demonstrates autonomous chemistry on Emerald Cloud Lab", "summary": "CMU researchers published Coscientist in Nature, showing an LLM system that designed, planned, and executed chemistry experiments through cloud-lab automation.", "company": "Coscientist (CMU)", "featured": false, "sourceIds": [ "cmu-coscientist-202312", "nature-coscientist-202312" ] }, { "company_id": "kebotix", "date": "2019-05", "title": "Kebotix raises Series A for self-driving materials discovery", "summary": "Kebotix raised $11.4M to scale an AI-powered self-driving lab for rapid chemicals and materials discovery.", "company": "Kebotix", "featured": false, "sourceIds": [ "finsmes-kebotix-series-a-202004" ] }, { "company_id": "citrine-informatics", "date": "2023-01", "title": "Citrine raises Series C for materials AI platform", "summary": "Citrine closed $16M to expand its AI-driven platform for materials, chemicals, and manufactured-product development.", "company": "Citrine Informatics", "featured": false, "sourceIds": [ "businesswire-citrine-series-c-202301" ] }, { "company_id": "artificial", "date": "2021-05", "title": "Artificial raises Series A for lab orchestration", "summary": "Artificial raised $21.5M from Microsoft M12 and others to build aLab Suite for life-sciences lab automation and orchestration.", "company": "Artificial", "featured": false, "sourceIds": [ "techcrunch-artificial-series-a-202105" ] }, { "company_id": "synthace", "date": "2021-11", "title": "Synthace raises Series C for experiment automation software", "summary": "Synthace raised $35M to expand its R&D cloud for designing experiments, generating automation instructions, and preserving structured data.", "company": "Synthace", "featured": false, "sourceIds": [ "finsmes-synthace-series-c-202111" ] }, { "company_id": "tetrascience", "date": "2021-04", "title": "TetraScience raises Series B for R&D Data Cloud", "summary": "TetraScience raised $80M for its scientific data cloud, supporting centralized and AI-ready experimental data across enterprise R&D.", "company": "TetraScience", "featured": false, "sourceIds": [ "fierce-tetrascience-series-b-202104" ] }, { "company_id": "opentrons", "date": "2023-05", "title": "Opentrons launches Flex lab robot", "summary": "Opentrons launched Flex, an open and modular liquid-handling robot positioned for reproducible, programmable, and AI-compatible lab automation.", "company": "Opentrons", "featured": false, "sourceIds": [ "official-opentrons-flex", "techcrunch-opentrons-flex-202305" ] }, { "company_id": "generate-biomedicines", "date": "2023-09", "title": "Generate Biomedicines raises Series C for generative biology pipeline", "summary": "Generate raised $273M to advance preclinical and clinical protein therapeutics built from its machine-learning-powered platform.", "company": "Generate Biomedicines", "featured": false, "sourceIds": [ "businesswire-generate-series-c-202309" ] }, { "company_id": "iambic-therapeutics", "date": "2023-10", "title": "Iambic raises Series B and announces NVIDIA collaboration", "summary": "Iambic raised $100M and paired its AI-driven drug discovery platform with NVIDIA infrastructure as it moved candidates toward the clinic.", "company": "Iambic Therapeutics", "featured": false, "sourceIds": [ "fierce-iambic-series-b-202310" ] }, { "company_id": "genesis-molecular-ai", "date": "2023-08", "title": "Genesis raises Series B for GEMS drug discovery platform", "summary": "Genesis Therapeutics raised $200M to advance AI-enabled drug programs and expand its GEMS molecular-design platform.", "company": "Genesis Molecular AI", "featured": false, "sourceIds": [ "fierce-genesis-series-b-202308", "cen-genesis-series-b-202308" ] }, { "company_id": "argonne-national-laboratory", "date": "2023-08", "title": "Argonne formalizes Autonomous Discovery initiative", "summary": "Argonne described an autonomous-discovery program spanning AI, robotics, self-driving chemistry, data platforms, and automated scientific decision loops.", "company": "Argonne National Laboratory", "featured": false, "sourceIds": [ "official-argonne-autonomous-discovery", "official-argonne-science101-autonomous-discovery" ] }, { "company_id": "oak-ridge-national-laboratory", "date": "2024-06", "title": "ORNL demonstrates AI-controlled autonomous materials synthesis", "summary": "ORNL reported an autonomous pulsed-laser-deposition tool where AI analyzed material quality and controlled the next experiment without human supervision.", "company": "Oak Ridge National Laboratory", "featured": false, "sourceIds": [ "official-ornl-autonomous-science" ] }, { "company_id": "lawrence-livermore-national-laboratory", "date": "2025-07", "title": "LLNL announces APEX self-driving alloy lab", "summary": "LLNL detailed APEX, a platform intended to autonomously design, fabricate, prepare, and characterize 3D-printed alloy samples.", "company": "Lawrence Livermore National Laboratory", "featured": false, "sourceIds": [ "official-llnl-apex-202507", "str-llnl-apex-202601" ] }, { "company_id": "nist", "date": "2020-11", "title": "NIST CAMEO autonomously discovers a new material", "summary": "NIST researchers showed CAMEO, an AI-driven closed-loop method that discovered a new material while reducing trial-and-error lab work.", "company": "NIST", "featured": false, "sourceIds": [ "nist-autonomous-systems-materials-202009" ] }, { "company_id": "nrel", "date": "2024-10", "title": "NREL publishes text-to-test control software workflow", "summary": "NREL researchers showed how LLMs can generate control software for materials-science instruments, supporting more accessible autonomous experimentation.", "company": "NREL", "featured": false, "sourceIds": [ "official-nrel-autonomous-experimentation", "rsc-text-to-test-202410" ] }, { "company_id": "brookhaven-national-laboratory", "date": "2023-01", "title": "Brookhaven autonomous methods discover new nanostructures", "summary": "Brookhaven researchers demonstrated AI-driven autonomous methods that discovered three new nanostructures at the Center for Functional Nanomaterials.", "company": "Brookhaven National Laboratory", "featured": false, "sourceIds": [ "sciencedaily-bnl-ai-nanostructures-202301" ] }, { "company_id": "materials-genome-initiative", "date": "2026-04", "title": "MGI releases autonomous experimentation materials R&D report", "summary": "The Materials Genome Initiative published an autonomous experimentation report summarizing AMII workshop findings and next steps for accelerated materials R&D.", "company": "Materials Genome Initiative", "featured": false, "sourceIds": [ "official-mgi-autonomous-experimentation-report" ] }, { "company_id": "chai-discovery", "date": "2025-06", "title": "Chai-2 reports zero-shot antibody design with double-digit hit rates", "summary": "Chai Discovery unveiled Chai-2 as a zero-shot antibody and protein-binder design platform, reporting double-digit experimental success rates, fewer than 20 experimental designs per target, and more than a 100-fold improvement over previous computational methods.", "company": "Chai Discovery", "featured": false, "sourceIds": [ "businesswire-chai2-202506", "biorxiv-chai2-zero-shot-202507" ] }, { "company_id": "chai-discovery", "date": "2025-12", "title": "Chai Discovery raises $130M Series B at $1.3B valuation", "summary": "Chai raised a $130M Series B co-led by Oak HC/FT and General Catalyst, bringing total funding above $225M and valuing the company at $1.3B.", "company": "Chai Discovery", "featured": false, "sourceIds": [ "businesswire-chai-series-b-202512", "techcrunch-chai-series-b-202512" ] }, { "company_id": "chai-discovery", "date": "2026-01", "title": "Chai and Lilly collaborate on AI biologics discovery", "summary": "Chai announced a collaboration with Eli Lilly to accelerate biologics discovery, including a custom model trained on proprietary Lilly data and deployment of Chai's frontier AI platform in Lilly workflows.", "company": "Chai Discovery", "featured": false, "sourceIds": [ "biospace-chai-lilly-202601" ] }, { "company_id": "ai2", "date": "2026-04", "title": "Ai2 updates AstaBench scientific-agent leaderboard", "summary": "Ai2's spring 2026 AstaBench update reported new frontier-model results on a suite of 2,400+ scientific research problems spanning literature understanding, coding and execution, data analysis, and end-to-end discovery.", "company": "Ai2", "featured": false, "sourceIds": [ "official-astabench", "ai2-astabench-update-202604" ] }, { "company_id": "robochem-flex", "date": "2026-04", "title": "RoboChem-Flex publishes affordable modular self-driving lab", "summary": "Nature Synthesis published RoboChem-Flex, a low-cost modular self-driving laboratory for autonomous reaction optimization with Python software, Bayesian optimization, closed-loop and human-in-the-loop modes, and six validation case studies.", "company": "RoboChem-Flex", "featured": false, "sourceIds": [ "nature-robochem-flex-202604" ] }, { "company_id": "converge-bio", "date": "2026-01", "title": "Converge Bio raises $25M Series A", "summary": "Converge Bio raised a $25M Series A led by Bessemer Venture Partners, bringing total funding to $30M and reporting over a dozen pharma and biotech customers across target discovery, antibody design, and protein manufacturing optimization.", "company": "Converge Bio", "featured": false, "sourceIds": [ "official-converge-series-a-202601", "techcrunch-converge-series-a-202601" ] }, { "company_id": "generare", "date": "2026-04", "title": "Generare raises €20M Series A for microbial chemistry platform", "summary": "Generare raised a €20M Series A co-led by Alven and Daphni to scale its platform for reading microbial genomes, expressing silent chemistry, and generating proprietary novel small-molecule data.", "company": "Generare", "featured": false, "sourceIds": [ "alven-generare-series-a-202604", "biologydigital-generare-series-a-202604" ] }, { "company_id": "alloy-therapeutics", "date": "2026-04", "title": "Alloy Therapeutics raises $40M Series E at $1B valuation", "summary": "Alloy raised a $40M Series E at a $1B valuation, reporting 200+ partners, 100+ licensed therapeutic programs, 22 clinical programs, and positioning its AI, real-world-data, and wet-lab capabilities as on-demand biotech infrastructure.", "company": "Alloy Therapeutics", "featured": false, "sourceIds": [ "businesswire-alloy-series-e-202604", "citybiz-alloy-series-e-202604" ] }, { "company_id": "scivity-labs", "date": "2026-01", "title": "Scivity Labs begins", "summary": "Scivity was founded in 2025 to build AI-native research infrastructure for autonomous computational research workflows.", "company": "Scivity Labs", "featured": false, "sourceIds": [ "official-scivity-home-202605", "official-scivity-changelog-202604" ] }, { "company_id": "scivity-labs", "date": "2026-03", "title": "Scivity publishes its verification surface", "summary": "Scivity made verification a public product surface, describing validation outcomes while keeping thresholds, calibration, and check-composition logic private.", "company": "Scivity Labs", "featured": false, "sourceIds": [ "official-scivity-verification-202603", "official-scivity-changelog-202604" ] }, { "company_id": "scivity-labs", "date": "2026-04", "title": "Scivity discloses first end-to-end autonomous research run", "summary": "Scivity reported an autonomous research run spanning experiment design, execution, verification, and report generation without human intervention.", "company": "Scivity Labs", "featured": false, "sourceIds": [ "official-scivity-changelog-202604" ] }, { "company_id": "tetsuwan-scientific", "date": "2024-11", "title": "Tetsuwan reports first autonomous AI-scientist deployment in a rare-disease lab", "summary": "Tetsuwan said its inaugural autonomous AI system was operational in a rare-disease gene-therapeutics lab, pairing the deployment with a $2.7M pre-seed to expand ResearchOS and the cloud-lab roadmap.", "company": "Tetsuwan Scientific", "featured": false, "sourceIds": [ "synbiobeta-tetsuwan-preseed-202411", "techcrunch-tetsuwan-preseed-202412" ] }, { "company_id": "microsoft-discovery", "date": "2025-05", "title": "Microsoft launches Discovery at Build 2025", "summary": "Microsoft introduced Discovery as an enterprise agentic R&D platform combining a graph-based knowledge engine, hypothesis generation, simulation, and iterative analysis on Azure.", "company": "Microsoft Discovery", "featured": false, "sourceIds": [ "official-microsoft-discovery-intro-202505", "techcrunch-microsoft-discovery-202505" ] }, { "company_id": "microsoft-discovery", "date": "2026-04", "title": "Microsoft expands Discovery preview with enterprise R&D partners", "summary": "Microsoft expanded preview access for Discovery, adding partner integrations, customer examples, and explicit interoperability with physical labs, robotics, and HPC-backed R&D workflows.", "company": "Microsoft Discovery", "featured": false, "sourceIds": [ "official-microsoft-discovery-solution-202604", "official-microsoft-discovery-preview-202604" ] }, { "company_id": "edison-scientific", "date": "2026-02", "title": "Edison introduces PaperQA3 and upgrades Edison Literature", "summary": "Edison launched PaperQA3 as the new algorithm behind Edison Literature and Kosmos literature search, adding multimodal figure-and-table reasoning across 150M+ papers and patents.", "company": "Edison Scientific", "featured": false, "sourceIds": [ "official-edison-paperqa3-202602" ] }, { "company_id": "edison-scientific", "date": "2026-03", "title": "Edison deepens NVIDIA partnership around Kosmos and BixBench-Hypothesis", "summary": "Edison described NVIDIA as a core partner for scaling Kosmos, multimodal literature understanding, and the new BixBench-Hypothesis training and evaluation stack for scientific agents.", "company": "Edison Scientific", "featured": false, "sourceIds": [ "official-edison-nvidia-202603", "nvidia-edison-nemotron-case-study-202603" ] }, { "company_id": "ai2", "date": "2025-08", "title": "Ai2 launches Asta open agentic ecosystem for science", "summary": "Ai2 introduced Asta, an open agentic ecosystem for science covering scholarly research assistants, the AstaBench benchmark suite (2,400+ problems across 11 benchmarks), and a developer toolkit with open-source agents, post-trained science LMs, and a 200M-paper Scientific Corpus Tool.", "company": "Ai2", "featured": false, "sourceIds": [ "official-ai2-asta-launch-202508", "businesswire-ai2-asta-202508", "ai2-astabench-blog-202508" ] }, { "company_id": "anthropic", "date": "2026-04", "title": "Anthropic releases BioMysteryBench for bioinformatics research", "summary": "Anthropic introduced BioMysteryBench, a 99-task bioinformatics evaluation with expert baselines and reliability analysis, alongside Claude evaluations on the dataset.", "company": "Anthropic", "featured": false, "sourceIds": [ "official-anthropic-biomysterybench-202604", "huggingface-biomysterybench-202604" ] }, { "company_id": "amazon-bio-discovery", "date": "2026-04", "title": "AWS launches Amazon Bio Discovery agentic drug-research platform", "summary": "AWS announced Amazon Bio Discovery on April 28: agentic AI on Bedrock that selects from 40+ biology foundation models, routes candidates to integrated CRO and cloud-lab partners (Ginkgo Nebula, Twist Bioscience, A-Alpha Bio anticipated) for synthesis and testing, and returns results into a lab-in-the-loop cycle. Early Memorial Sloan Kettering project triaged 300,000 antibody candidates in weeks.", "company": "Amazon Bio Discovery", "featured": false, "sourceIds": [ "official-aws-bio-discovery-blog-202604", "genengnews-aws-bio-discovery-202604", "aipedia-aws-bio-discovery-202604" ] }, { "company_id": "profluent", "date": "2025-11", "title": "Profluent closes $106M Series B for programmable biology", "summary": "Profluent raised a $106M Series B co-led by Altimeter Capital and Bezos Expeditions, taking total funding to $150M, to scale frontier protein language models and OpenCRISPR-1 adoption across pharma and academic users.", "company": "Profluent", "featured": false, "sourceIds": [ "businesswire-profluent-series-b-202511" ] }, { "company_id": "profluent", "date": "2026-05", "title": "Eli Lilly / Profluent strike $2.25B AI recombinase pact", "summary": "Eli Lilly and Profluent announced a multi-program strategic research collaboration to design and commercialize custom site-specific recombinases for severe-unmet-need diseases. Deal carries an upfront payment, committed R&D funding, up to $2.25B in development and commercial milestones, and tiered royalties on net sales.", "company": "Profluent", "featured": false, "sourceIds": [ "biospace-profluent-lilly-202604", "airstreet-profluent-lilly-202604" ] } ], "benchmarks": [ { "name": "Artificial Analysis Intelligence Index", "domain": "General model intelligence", "owner": "Artificial Analysis", "metric": "Composite score across agents, coding, science, reasoning, knowledge, and instruction following", "topResult": "Intelligence Index v4.0 aggregates ten evaluations (GPQA Diamond, Humanity's Last Exam, SciCode, Terminal-Bench Hard, GDPval-AA and more); Claude Opus 4.8 leads at 61, ahead of GPT-5.5 (60) and Gemini 3.1 Pro (57).", "resultDate": "2026-05", "relevanceToScience": "Provides a production-oriented view of which frontier models are strong enough to act as reasoning engines inside scientific agents.", "sourceIds": [ "artificial-analysis-leaderboard", "artificial-analysis-methodology", "artificial-analysis-intelligence-index", "axios-anthropic-opus48-202605" ], "benchmark_id": "artificial-analysis-intelligence-index" }, { "name": "Epoch AI Capabilities", "domain": "Capability trends", "owner": "Epoch AI", "metric": "Benchmark results across 40+ evaluations, with internal and external result provenance", "topResult": "On the Epoch Capabilities Index, GPT-5.5 Pro holds the top score at 159 (90% CI 156-162), on a scale calibrated so GPT-5 = 150 and Claude 3.5 Sonnet = 130.", "resultDate": "2026-05", "relevanceToScience": "Useful for seeing whether scientific reasoning, agentic work, math, coding, and multimodal capabilities are improving fast enough to change lab workflows.", "sourceIds": [ "epoch-capabilities", "epoch-benchmarking-about", "epoch-capabilities-index", "epoch-eci-gpt55pro-202604" ], "benchmark_id": "epoch-ai-capabilities" }, { "name": "FrontierMath", "domain": "Mathematics", "owner": "Epoch AI", "metric": "Accuracy on extremely difficult math problems and open-problem variants", "topResult": "GPT-5.5 Pro set a new high, reaching 52% on Tiers 1-3 and 40% on the research-grade Tier 4 — a step up from prior frontier models, though the hardest tier remains far from solved.", "resultDate": "2026-06", "relevanceToScience": "High-end mathematical reasoning is one of the cleanest proxies for whether models can contribute to formal scientific discovery.", "sourceIds": [ "epoch-capabilities", "epoch-benchmarking-about", "epoch-eci-gpt55pro-202604" ], "benchmark_id": "frontiermath" }, { "name": "GPQA Diamond", "domain": "Scientific reasoning", "owner": "GPQA / tracked by Epoch AI and Artificial Analysis", "metric": "Expert-level graduate science multiple-choice accuracy", "topResult": "Now near saturation: leaders cluster around 94% (Gemini 3.1 Pro ~94.3%, Claude Opus 4.7 ~94.2%), above the ~81% expert-validator baseline, so most of the headroom has closed.", "resultDate": "2026-05", "relevanceToScience": "Directly probes PhD-level physics, chemistry, and biology reasoning; with leaders near 94% its remaining signal is mostly about reliability and contamination rather than raw knowledge.", "sourceIds": [ "epoch-capabilities", "artificial-analysis-methodology", "stanford-ai-index-2026", "artificial-analysis-intelligence-index" ], "benchmark_id": "gpqa-diamond" }, { "name": "SciCode", "domain": "Scientific coding", "owner": "Scientific coding benchmark / tracked by Artificial Analysis", "metric": "Pass rate on scientific programming tasks", "topResult": "Retained as the scientific-coding component of Artificial Analysis Intelligence Index v4.0, where it stays unsaturated even as general coding benchmarks climb toward the ceiling.", "resultDate": "2026-05", "relevanceToScience": "Measures whether models can turn scientific specifications into executable code, a core dependency for autonomous analysis and simulation.", "sourceIds": [ "artificial-analysis-methodology", "artificial-analysis-intelligence-index" ], "benchmark_id": "scicode" }, { "name": "SWE-bench Verified", "domain": "Agentic software engineering", "owner": "SWE-bench / tracked by Epoch AI", "metric": "Resolved real GitHub issues", "topResult": "The restricted Claude Mythos Preview reports 93.9%, well above the generally available cluster at ~80-81% (Claude Opus 4.5 80.9%, Opus 4.6 80.8%, Gemini 3.1 Pro 80.6%); training-data contamination on this split motivated the harder, contamination-resistant SWE-bench Pro.", "resultDate": "2026-06", "relevanceToScience": "Software-engineering agents are a leading indicator for whether models can operate long-horizon scientific toolchains and repair failed experiments.", "sourceIds": [ "epoch-capabilities", "epoch-benchmarking-about", "swebench-overview", "codeant-swebench-leaderboard-202602", "anthropic-claude-mythos-202604" ], "benchmark_id": "swe-bench-verified" }, { "name": "APEX-Agents", "domain": "Long-horizon agents", "owner": "Mercor / tracked by Epoch AI and Artificial Analysis", "metric": "Pass@1 task completion in realistic multi-application workflows", "topResult": "Added to the Epoch Capabilities Index in 2026", "resultDate": "2026-03", "relevanceToScience": "Lab automation requires agents that coordinate files, tools, state, and multi-step objectives rather than answering isolated prompts.", "sourceIds": [ "epoch-capabilities", "artificial-analysis-leaderboard" ], "benchmark_id": "apex-agents" }, { "name": "Humanity's Last Exam", "domain": "Cross-domain expert reasoning", "owner": "HLE / tracked by Epoch AI and Artificial Analysis", "metric": "Accuracy on hard expert-written questions", "topResult": "Still far from saturated: leaders sit at ~34-38% (Gemini 3 Pro Preview ~37.5%, Claude Opus 4.6 Thinking ~34.4%), so it remains one of the few knowledge benchmarks with real headroom.", "resultDate": "2026-05", "relevanceToScience": "A broad stress test for frontier models, useful only when interpreted alongside domain-specific science benchmarks and tool-use evaluations.", "sourceIds": [ "epoch-capabilities", "artificial-analysis-methodology", "nature-hle", "artificial-analysis-intelligence-index" ], "benchmark_id": "humanitys-last-exam" }, { "name": "ARC-AGI-2", "domain": "Fluid reasoning and generalization", "owner": "ARC Prize Foundation (Francois Chollet)", "metric": "Accuracy on novel abstraction-and-reasoning tasks built to resist memorization", "topResult": "GPT-5.5 leads at ~85%, with Gemini 3.1 Pro ~77% and Claude Opus 4.6 ~69% — a steep climb from near-zero in 2024, though still short of the saturated knowledge benchmarks.", "resultDate": "2026-05", "relevanceToScience": "Tests whether models can solve genuinely new problems from a few examples rather than recall training data — the capability autonomous discovery depends on most.", "sourceIds": [ "arcprize-arc-agi2-leaderboard-202605", "llm-stats-arc-agi2-202605" ], "benchmark_id": "arc-agi-2" }, { "name": "METR Task-Completion Time Horizons", "domain": "Long-horizon agency", "owner": "METR", "metric": "Length of software/ML task, measured in human-expert time, that an agent completes at 50% and 80% reliability", "topResult": "Claude Mythos Preview (added May 8, 2026) pushes the 50% time horizon past ~16 hours — the ceiling of METR's reliable measurement — while Claude Opus 4.6 and GPT-5.2 cluster around 5-6 hours. The post-2023 doubling time is about four to five months (~130 days under TH1.1) and accelerating.", "resultDate": "2026-06", "relevanceToScience": "Long-horizon reliability is the binding constraint for running multi-step experiments and analyses without human supervision, making this the clearest single signal for the agency dimension.", "sourceIds": [ "metr-time-horizons-current", "metr-mythos-time-horizon-202605", "anthropic-claude-mythos-202604" ], "benchmark_id": "metr-time-horizons" }, { "name": "SWE-bench Pro", "domain": "Agentic software engineering (contamination-resistant)", "owner": "Scale AI SEAL / SWE-bench Pro", "metric": "Resolved real, multi-language GitHub issues on held-out tasks designed to resist training-data contamination", "topResult": "The restricted Claude Mythos Preview reports 77.8%, far above the generally available cluster at ~44-46% (Claude Opus 4.5 45.9%, Claude Sonnet 4.5 43.6%, Gemini 3 Pro 43.3%); the gap to the ~81% Verified split shows how much Verified was inflated by contamination.", "resultDate": "2026-06", "relevanceToScience": "A cleaner read than SWE-bench Verified on whether agents can maintain and repair real scientific toolchains rather than reproduce memorized solutions.", "sourceIds": [ "morphllm-swebench-pro-202603", "codeant-swebench-leaderboard-202602", "anthropic-claude-mythos-202604" ], "benchmark_id": "swe-bench-pro" }, { "name": "Terminal-Bench", "domain": "Computer-use and tool agents", "owner": "Terminal-Bench / tracked by Stanford AI Index and Artificial Analysis", "metric": "Success rate on real terminal and command-line tasks completed end-to-end by an agent", "topResult": "Agent success on real-world terminal tasks rose from about 20% in 2025 to 77.3% in the 2026 AI Index; on Terminal-Bench 2.1 the new Gemini 3.5 Flash reaches 76.2% (vs 70.3% for Gemini 3.1 Pro) at roughly 4x output speed, and the harder Terminal-Bench Hard split is part of the Artificial Analysis Intelligence Index.", "resultDate": "2026-06", "relevanceToScience": "Command-line and tool fluency is the substrate for running analyses, managing data, and operating instruments, so it tracks the practical execution side of autonomy.", "sourceIds": [ "stanford-ai-index-2026", "artificial-analysis-intelligence-index", "google-gemini35-flash-202605" ], "benchmark_id": "terminal-bench" }, { "name": "BixBench", "domain": "Bioinformatics agents", "owner": "FutureHouse / ScienceMachine", "metric": "Open-answer accuracy on real-world bioinformatics analysis scenarios", "topResult": "Public benchmark with 53 analysis scenarios and 296 questions for agentic computational-biology workflows", "resultDate": "2025-03", "relevanceToScience": "Measures whether agents can explore biological datasets, run multi-step analyses, and interpret results rather than only answer static science questions.", "sourceIds": [ "official-futurehouse-bixbench-202503", "arxiv-bixbench-202503" ], "benchmark_id": "bixbench" }, { "name": "LABBench2", "domain": "Biology research tasks", "owner": "Edison Scientific", "metric": "Performance across 1,892 practical biology-research tasks spanning literature, databases, sequences, protocols, patents, trials, and source quality", "topResult": "Open dataset and harness with published model comparisons across 11 task families", "resultDate": "2026-05", "relevanceToScience": "Moves biology-agent evaluation toward practical research work, retrieval, file handling, and tool use rather than short-form knowledge recall.", "sourceIds": [ "official-edison-labbench2-202605", "huggingface-labbench2-202603" ], "benchmark_id": "labbench2" }, { "name": "BioMysteryBench", "domain": "Bioinformatics agents", "owner": "Anthropic", "metric": "Accuracy and reliability on 99 expert-level bioinformatics tasks with objective ground truth from real experimental data", "topResult": "Anthropic reports Claude-family and expert baselines, including separate human-solvable and human-difficult task sets", "resultDate": "2026-04", "relevanceToScience": "Tests whether frontier agents can produce reproducible scientific conclusions from messy biological data, including problems not solved by expert panels.", "sourceIds": [ "official-anthropic-biomysterybench-202604", "huggingface-biomysterybench-202604" ], "benchmark_id": "biomysterybench" }, { "name": "GeneBench", "domain": "Genomics and quantitative biology agents", "owner": "OpenAI", "metric": "Pass rate on 103 multi-stage scientific data analysis tasks across 10 genomics and quantitative biology domains", "topResult": "Frontier models score roughly 25-33% on tasks that often correspond to multi-day projects for expert computational biologists", "resultDate": "2026-04", "relevanceToScience": "Probes whether agents can clean assay or clinical data, run exploratory analysis, select statistical models, and produce conclusions that inform downstream scientific decisions.", "sourceIds": [ "openai-genebench-202604", "biorxiv-genebench-202604" ], "benchmark_id": "genebench" }, { "benchmark_id": "lab-bench", "name": "LAB-Bench", "domain": "biology", "owner": "FutureHouse / LAB-Bench authors", "metric": "task accuracy and open-response accuracy", "topResult": "Human experts 0.70-1.00 across many subtasks; frontier models remain uneven.", "resultDate": "2024-07", "relevanceToScience": "Measures biology research-agent skills across literature QA, table/figure reasoning, protocols, databases, sequences, and cloning scenarios.", "sourceIds": [ "arxiv-labbench-202407" ] }, { "benchmark_id": "litqa2-paperqa", "name": "LitQA2 / PaperQA2", "domain": "scientific literature", "owner": "FutureHouse / PaperQA authors", "metric": "precision, accuracy, DOI recall, contradiction-detection AUC", "topResult": "PaperQA2 precision 85.2% and accuracy 66.0% on LitQA2.", "resultDate": "2024-09", "relevanceToScience": "Evaluates retrieval-grounded scientific literature QA, synthesis, and contradiction detection against human experts.", "sourceIds": [ "arxiv-paperqa2-202409" ] }, { "benchmark_id": "foldbench", "name": "FoldBench", "domain": "biology", "owner": "BEAM-Labs FoldBench authors", "metric": "DockQ AUC, DockQ success rate, LDDT, ligand success rate", "topResult": "AlphaFold 3 leads most measured all-atom structure-prediction tasks.", "resultDate": "2025-12", "relevanceToScience": "Independent benchmark of all-atom biomolecular structure prediction across protein-ligand, protein-protein, antibody, and nucleic-acid tasks.", "sourceIds": [ "nature-foldbench-2025" ] }, { "benchmark_id": "gnome-materials-discovery", "name": "GNoME materials discovery benchmark", "domain": "materials", "owner": "Google DeepMind", "metric": "new stable crystal structures on updated convex hull", "topResult": "GNoME reports 381,000 newly stable convex-hull entries from 2.2M candidate structures.", "resultDate": "2023-11", "relevanceToScience": "Measures AI-assisted inorganic materials discovery at the scale of DFT-verified stable structures.", "sourceIds": [ "nature-gnome-202311" ] }, { "benchmark_id": "alphaevolve-algorithm-discovery", "name": "AlphaEvolve algorithm discovery tasks", "domain": "algorithmic science", "owner": "Google DeepMind", "metric": "objective-specific best-known construction score", "topResult": "AlphaEvolve found a 48-multiplication 4x4 complex matrix multiplication algorithm versus Strassen 49.", "resultDate": "2025-05", "relevanceToScience": "Measures autonomous code-evolution workflows on verifiable mathematical and scientific optimization objectives.", "sourceIds": [ "official-alphaevolve-202505", "arxiv-alphaevolve-202506" ] }, { "benchmark_id": "alphagenome-regulatory-benchmarks", "name": "AlphaGenome regulatory genomics benchmarks", "domain": "genomics", "owner": "Google DeepMind", "metric": "benchmark tasks with state-of-the-art performance", "topResult": "AlphaGenome achieved SOTA on 22 of 24 track-prediction tasks and 25 of 26 variant-effect tasks.", "resultDate": "2025-06", "relevanceToScience": "Aggregates genome-track prediction and regulatory variant-effect prediction measurements against external genomics baselines.", "sourceIds": [ "nature-alphagenome-202506" ] }, { "benchmark_id": "cfps-autonomous-lab-optimization", "name": "CFPS autonomous-lab optimization", "domain": "synthetic biology", "owner": "OpenAI / Ginkgo Bioworks", "metric": "sfGFP production cost and titer improvement under closed-loop cell-free protein synthesis optimization", "topResult": "GPT-5 + Ginkgo Cloud Lab reported $422/g versus a $698/g state-of-the-art CFPS benchmark, plus a 27% titer increase after 36,000+ unique reactions.", "resultDate": "2026-02", "relevanceToScience": "Measures an AI model plus execution-layer autonomous lab loop on a wet-lab optimization target with explicit cost, scale, and titer outcomes.", "sourceIds": [ "openai-ginkgo-cfps-autonomous-lab-202602", "ginkgo-openai-cfps-202602", "openai-ginkgo-cfps-preprint-202602", "biorxiv-openai-ginkgo-cfps-202602" ] }, { "benchmark_id": "astabench", "name": "AstaBench", "domain": "scientific agents", "owner": "Ai2", "metric": "overall scientific-agent task accuracy and cost across literature understanding, coding/execution, data analysis, and end-to-end discovery", "topResult": "Spring 2026 update reports Claude Opus 4.7 ReAct at 58.0% overall, GPT-5.5 ReAct at 52.9%, and GPT-5.4 ReAct at 46.5%.", "resultDate": "2026-04", "relevanceToScience": "AstaBench directly evaluates whether agents can perform grounded scientific research workflows rather than isolated question answering.", "sourceIds": [ "official-astabench", "ai2-astabench-blog-202508", "ai2-astabench-update-202604" ] }, { "benchmark_id": "fire-bench", "name": "FIRE-Bench", "domain": "Autonomous scientific discovery", "owner": "FIRE-Bench authors (arXiv)", "metric": "Rediscovery success when an agent is given only a high-level research question and must explore ideas, design experiments, implement and execute code, and derive conclusions", "topResult": "Even the strongest agent systems achieve only limited rediscovery success, showing full-cycle scientific research remains hard for current models.", "resultDate": "2026-02", "relevanceToScience": "FIRE-Bench is the closest public proxy for end-to-end autonomous science: not answering questions, but running the full loop from question to verified conclusion — exactly the workflow autonomous labs must own.", "sourceIds": [ "arxiv-firebench-202602" ] }, { "benchmark_id": "autoresearchbench", "name": "AutoResearchBench", "domain": "Scientific literature agents", "owner": "AutoResearchBench authors (arXiv)", "metric": "Accuracy on complex literature-discovery tasks requiring corpus search, in-depth full-paper reading, fine-grained technical verification, and a stopping decision", "topResult": "Public benchmark probing whether an agent can search a large, up-to-date corpus, read papers in depth, verify fine-grained conditions, and decide when its search is complete.", "resultDate": "2026-04", "relevanceToScience": "Literature grounding is the entry point to autonomous research; this measures whether agents can build a correct, verified evidence base before acting rather than stopping at shallow retrieval.", "sourceIds": [ "arxiv-autoresearchbench-202604" ] } ], "research-outputs": [ { "output_id": "alab-nature-2023", "output_type": "paper", "doi": "10.1038/s41586-023-06734-w", "arxiv_id": null, "biorxiv_id": null, "pubmed_id": "38030721", "title": "An autonomous laboratory for the accelerated synthesis of inorganic materials", "date": "2023-11-29", "venue": "Nature", "peer_reviewed": true, "open_access": true, "license": "CC BY 4.0", "code_available": null, "dataset_available": true, "model_available": null, "source_tier": "A", "claim_status": "corrected", "last_verified": "2026-05-10", "source_ids": [ "nature-alab-202312", "nature-alab-correction-202601" ], "entity_ids": [ "a-lab-lbnl" ], "entity_names": [ "A-Lab (LBNL)" ] }, { "output_id": "coscientist-nature-2023", "output_type": "paper", "doi": "10.1038/s41586-023-06792-0", "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "title": "Autonomous chemical research with large language models", "date": "2023-12-20", "venue": "Nature", "peer_reviewed": true, "open_access": true, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-10", "source_ids": [ "nature-coscientist-202312" ], "entity_ids": [ "coscientist-cmu" ], "entity_names": [ "Coscientist (CMU)" ] }, { "output_id": "futurehouse-disco-2026", "output_type": "preprint", "doi": null, "arxiv_id": "2604.05181", "biorxiv_id": null, "pubmed_id": null, "title": "General Multimodal Protein Design Enables DNA-Encoding of Chemistry", "date": "2026-04-06", "venue": "arXiv", "peer_reviewed": false, "open_access": true, "license": null, "code_available": null, "dataset_available": null, "model_available": true, "source_tier": "B", "claim_status": "reported", "last_verified": "2026-05-10", "source_ids": [ "arxiv-disco-enzyme-design-202604", "official-futurehouse-disco-202604" ], "entity_ids": [ "futurehouse" ], "entity_names": [ "FutureHouse" ] }, { "output_id": "nrel-text-to-test-2024", "output_type": "paper", "doi": "10.1039/D4DD00143E", "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "title": "From text to test: AI-generated control software for materials science instruments", "date": "2024-10-23", "venue": "Digital Discovery", "peer_reviewed": true, "open_access": true, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-10", "source_ids": [ "rsc-text-to-test-202410" ], "entity_ids": [ "nrel" ], "entity_names": [ "NREL" ] }, { "doi": "10.1038/s41586-026-10265-5", "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": true, "dataset_available": true, "model_available": false, "last_verified": "2026-05-11", "output_id": "sakana-ai-scientist-nature-2026", "output_type": "paper", "title": "The AI Scientist: towards fully automated open-ended scientific discovery", "date": "2026-03-26", "venue": "Nature", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "confirmed", "source_ids": [ "nature-ai-scientist-202603" ], "entity_ids": [ "sakana-ai" ], "entity_names": [ "Sakana AI" ] }, { "doi": null, "arxiv_id": "2408.06292", "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": true, "dataset_available": true, "model_available": false, "last_verified": "2026-05-11", "output_id": "sakana-ai-scientist-v1-arxiv-2024", "output_type": "preprint", "title": "The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery", "date": "2024-08-12", "venue": "arXiv", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "arxiv-sakana-ai-scientist-202408" ], "entity_ids": [ "sakana-ai" ], "entity_names": [ "Sakana AI" ] }, { "doi": null, "arxiv_id": "2504.08066", "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": true, "dataset_available": true, "model_available": false, "last_verified": "2026-05-11", "output_id": "sakana-ai-scientist-v2-arxiv-2025", "output_type": "preprint", "title": "The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search", "date": "2025-04-10", "venue": "arXiv", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "arxiv-sakana-ai-scientist-v2-202504", "github-sakana-ai-scientist-v2-202504" ], "entity_ids": [ "sakana-ai" ], "entity_names": [ "Sakana AI" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": true, "last_verified": "2026-05-11", "output_id": "sakana-fugu-2026", "output_type": "technical_report", "title": "Fugu: A Multi-Agent Orchestration System as a Foundation Model", "date": "2026-04-01", "venue": "Sakana AI technical report", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-sakana-fugu-202604" ], "entity_ids": [ "sakana-ai" ], "entity_names": [ "Sakana AI" ] }, { "doi": null, "arxiv_id": "2409.13740", "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": true, "dataset_available": true, "model_available": true, "last_verified": "2026-05-11", "output_id": "futurehouse-paperqa2-2024", "output_type": "preprint", "title": "Language Agents Achieve Superhuman Synthesis of Scientific Knowledge", "date": "2024-09-20", "venue": "arXiv", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "arxiv-paperqa2-202409" ], "entity_ids": [ "futurehouse" ], "entity_names": [ "FutureHouse" ] }, { "doi": null, "arxiv_id": "2503.00096", "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": true, "model_available": false, "last_verified": "2026-05-11", "output_id": "futurehouse-bixbench-2025", "output_type": "preprint", "title": "BixBench: a Comprehensive Benchmark for LLM-based Agents in Computational Biology", "date": "2025-03-01", "venue": "arXiv", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "arxiv-bixbench-202503", "official-futurehouse-bixbench-202503" ], "entity_ids": [ "futurehouse" ], "entity_names": [ "FutureHouse" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": true, "model_available": false, "last_verified": "2026-05-11", "output_id": "edison-labbench2-2026", "output_type": "benchmark_dataset", "title": "LABBench2: An Improved Benchmark for Measuring AI in Biology Research", "date": "2026-04-13", "venue": "Edison Scientific / Hugging Face dataset", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-edison-labbench2-202605", "huggingface-labbench2-202603" ], "entity_ids": [ "edison-scientific" ], "entity_names": [ "Edison Scientific" ] }, { "doi": "10.1038/s41586-024-07487-w", "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": true, "dataset_available": true, "model_available": true, "last_verified": "2026-05-11", "output_id": "deepmind-alphafold3-nature-2024", "output_type": "paper", "title": "Accurate structure prediction of biomolecular interactions with AlphaFold 3", "date": "2024-05-08", "venue": "Nature", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "confirmed", "source_ids": [ "nature-alphafold3-202405" ], "entity_ids": [ "google-deepmind", "isomorphic-labs" ], "entity_names": [ "Google DeepMind", "Isomorphic Labs" ] }, { "doi": "10.1038/s41586-025-10014-0", "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": true, "dataset_available": true, "model_available": true, "last_verified": "2026-05-11", "output_id": "deepmind-alphagenome-nature-2025", "output_type": "paper", "title": "Advancing regulatory variant effect prediction with AlphaGenome", "date": "2025-06-25", "venue": "Nature", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "confirmed", "source_ids": [ "nature-alphagenome-202506" ], "entity_ids": [ "google-deepmind" ], "entity_names": [ "Google DeepMind" ] }, { "doi": "10.1038/s41586-023-06735-9", "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": true, "model_available": true, "last_verified": "2026-05-11", "output_id": "deepmind-gnome-nature-2023", "output_type": "paper", "title": "Scaling deep learning for materials discovery", "date": "2023-11-29", "venue": "Nature", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "confirmed", "source_ids": [ "nature-gnome-202311" ], "entity_ids": [ "google-deepmind" ], "entity_names": [ "Google DeepMind" ] }, { "doi": "10.1038/s41586-023-06924-6", "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "deepmind-funsearch-nature-2023", "output_type": "paper", "title": "Mathematical discoveries from program search with large language models", "date": "2023-12-14", "venue": "Nature", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "confirmed", "source_ids": [ "nature-funsearch-202312" ], "entity_ids": [ "google-deepmind" ], "entity_names": [ "Google DeepMind" ] }, { "doi": "10.1038/s41586-024-08252-9", "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": true, "dataset_available": true, "model_available": true, "last_verified": "2026-05-11", "output_id": "deepmind-gencast-nature-2024", "output_type": "paper", "title": "Probabilistic weather forecasting with machine learning", "date": "2024-12-04", "venue": "Nature", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "confirmed", "source_ids": [ "nature-gencast-202412" ], "entity_ids": [ "google-deepmind" ], "entity_names": [ "Google DeepMind" ] }, { "doi": null, "arxiv_id": "2506.13131", "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": false, "last_verified": "2026-05-11", "output_id": "deepmind-alphaevolve-2025", "output_type": "whitepaper", "title": "AlphaEvolve: A coding agent for scientific and algorithmic discovery", "date": "2025-05-14", "venue": "Google DeepMind / arXiv", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-alphaevolve-202505", "arxiv-alphaevolve-202506" ], "entity_ids": [ "google-deepmind" ], "entity_names": [ "Google DeepMind" ] }, { "doi": null, "arxiv_id": "2409.08022", "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": true, "last_verified": "2026-05-11", "output_id": "deepmind-alphaproteo-2024", "output_type": "technical_report", "title": "De novo design of high-affinity protein binders with AlphaProteo", "date": "2024-09-12", "venue": "arXiv", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "arxiv-alphaproteo-202409" ], "entity_ids": [ "google-deepmind" ], "entity_names": [ "Google DeepMind" ] }, { "doi": "10.1126/science.adg7492", "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": true, "model_available": true, "last_verified": "2026-05-11", "output_id": "deepmind-alphamissense-science-2023", "output_type": "paper", "title": "Accurate proteome-wide missense variant effect prediction with AlphaMissense", "date": "2023-09-19", "venue": "Science", "peer_reviewed": true, "open_access": false, "source_tier": "A", "claim_status": "confirmed", "source_ids": [ "science-alphamissense-202309" ], "entity_ids": [ "google-deepmind" ], "entity_names": [ "Google DeepMind" ] }, { "doi": "10.1038/s41467-025-67127-3", "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": true, "dataset_available": true, "model_available": false, "last_verified": "2026-05-11", "output_id": "foldbench-independent-2025", "output_type": "paper", "title": "FoldBench: systematic benchmark of all-atom biomolecular structure prediction models", "date": "2025-12-01", "venue": "Nature Communications", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "confirmed", "source_ids": [ "nature-foldbench-2025" ], "entity_ids": [ "google-deepmind" ], "entity_names": [ "Google DeepMind" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": true, "model_available": true, "last_verified": "2026-05-11", "output_id": "recursion-openphenom-s16-2024", "output_type": "model_release", "title": "OpenPhenom-S/16 foundation model for microscopy data", "date": "2024-11-12", "venue": "Google Cloud Model Garden / Recursion release", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-recursion-openphenom-202411" ], "entity_ids": [ "recursion-pharmaceuticals" ], "entity_names": [ "Recursion Pharmaceuticals" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": true, "dataset_available": true, "model_available": true, "last_verified": "2026-05-11", "output_id": "recursion-boltz2-2025", "output_type": "model_release", "title": "Boltz-2 biomolecular foundation model for structure and binding affinity prediction", "date": "2025-06-06", "venue": "MIT / Recursion release", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-recursion-boltz2-202506" ], "entity_ids": [ "recursion-pharmaceuticals" ], "entity_names": [ "Recursion Pharmaceuticals" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": true, "last_verified": "2026-05-11", "output_id": "recursion-phenom-brightfield-2024", "output_type": "technical_report", "title": "Phenomic foundation models for brightfield microscopy transfer", "date": "2024-09-11", "venue": "Recursion technical blog", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-recursion-brightfield-202409" ], "entity_ids": [ "recursion-pharmaceuticals" ], "entity_names": [ "Recursion Pharmaceuticals" ] }, { "doi": "10.1038/s41586-023-06728-8", "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": true, "dataset_available": true, "model_available": true, "last_verified": "2026-05-11", "output_id": "generate-chroma-nature-2023", "output_type": "paper", "title": "Illuminating protein space with a programmable generative model", "date": "2023-11-15", "venue": "Nature", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "confirmed", "source_ids": [ "nature-generate-chroma-202311", "official-generate-chroma-202311" ], "entity_ids": [ "generate-biomedicines" ], "entity_names": [ "Generate Biomedicines" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": true, "last_verified": "2026-05-11", "output_id": "iambic-neuralplexer-2024", "output_type": "paper", "title": "NeuralPLexer protein-ligand complex structure prediction", "date": "2024-02-12", "venue": "Nature Machine Intelligence", "peer_reviewed": true, "open_access": false, "source_tier": "A", "claim_status": "reported", "source_ids": [ "official-iambic-neuralplexer-202402" ], "entity_ids": [ "iambic-therapeutics" ], "entity_names": [ "Iambic Therapeutics" ] }, { "doi": "10.1038/s41467-025-66778-6", "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": true, "dataset_available": true, "model_available": true, "last_verified": "2026-05-11", "output_id": "insitro-posh-nature-communications-2025", "output_type": "paper", "title": "A pooled Cell Painting CRISPR screening platform enables de novo inference of gene function by self-supervised deep learning", "date": "2025-12-16", "venue": "Nature Communications", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "confirmed", "source_ids": [ "nature-comm-insitro-posh-202512", "official-insitro-posh-202512" ], "entity_ids": [ "insitro" ], "entity_names": [ "Insitro" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "argonne-autonomous-discovery-program-2026", "output_type": "program_report", "title": "Autonomous Discovery program and Science 101 explainer", "date": "2026-05-01", "venue": "Argonne National Laboratory", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-argonne-autonomous-discovery", "official-argonne-science101-autonomous-discovery" ], "entity_ids": [ "argonne-national-laboratory" ], "entity_names": [ "Argonne National Laboratory" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "ornl-autonomous-chemistry-lab-2026", "output_type": "program_report", "title": "Autonomous Chemistry Lab for accelerated materials discovery and innovation", "date": "2026-05-01", "venue": "Oak Ridge National Laboratory", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-ornl-autonomous-chemistry-lab", "official-ornl-autonomous-science" ], "entity_ids": [ "oak-ridge-national-laboratory" ], "entity_names": [ "Oak Ridge National Laboratory" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "llnl-apex-self-driving-lab-2025", "output_type": "program_report", "title": "APEX self-driving lab for accelerated alloy development", "date": "2025-07-01", "venue": "Lawrence Livermore National Laboratory", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-llnl-apex-202507", "str-llnl-apex-202601" ], "entity_ids": [ "lawrence-livermore-national-laboratory" ], "entity_names": [ "Lawrence Livermore National Laboratory" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "nist-autonomous-labs-program-2026", "output_type": "program_report", "title": "Autonomous laboratories and autonomous formulation lab programs", "date": "2026-05-01", "venue": "NIST", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-nist-autonomous-labs", "nist-autonomous-formulation-lab-202508", "nist-autonomous-systems-materials-202009" ], "entity_ids": [ "nist" ], "entity_names": [ "NIST" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "periodic-labs-platform-profile-202605", "output_type": "platform_report", "title": "Autonomous physical lab platform public technical profile", "date": "2026-05-11", "venue": "Periodic Labs", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-periodic-labs" ], "entity_ids": [ "periodic-labs" ], "entity_names": [ "Periodic Labs" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "lila-ai-science-factory-profile-202605", "output_type": "platform_report", "title": "AI Science Factory platform public technical profile", "date": "2026-05-11", "venue": "Lila Sciences", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-lila-sciences" ], "entity_ids": [ "lila-sciences" ], "entity_names": [ "Lila Sciences" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "autoscience-carl-agent-profile-202605", "output_type": "platform_report", "title": "Carl autonomous AI research agent public technical profile", "date": "2026-05-11", "venue": "Autoscience Institute", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-autoscience-carl", "official-autoscience-institute" ], "entity_ids": [ "autoscience-institute" ], "entity_names": [ "Autoscience Institute" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "autoscience-mira-production-research-202605", "output_type": "platform_report", "title": "Mira production research integration public technical profile", "date": "2026-05-11", "venue": "Autoscience Institute", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-autoscience-mira", "official-autoscience-institute" ], "entity_ids": [ "autoscience-institute" ], "entity_names": [ "Autoscience Institute" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "thesis-yc-profile-202605", "output_type": "platform_report", "title": "AI-native science tooling company profile", "date": "2026-05-11", "venue": "Y Combinator", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-thesis-platform", "official-thesis" ], "entity_ids": [ "thesis" ], "entity_names": [ "Thesis" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "zeon-systems-platform-profile-202605", "output_type": "platform_report", "title": "Zeon Systems AI science workflow platform profile", "date": "2026-05-11", "venue": "Zeon Systems", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-zeon-platform", "official-zeon-systems" ], "entity_ids": [ "zeon-systems" ], "entity_names": [ "Zeon Systems" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "automata-linq-platform-202605", "output_type": "platform_report", "title": "LINQ robotic lab automation platform technical profile", "date": "2026-05-11", "venue": "Automata", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-automata-linq", "official-automata" ], "entity_ids": [ "automata" ], "entity_names": [ "Automata" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "emerald-cloud-lab-hplc-anomaly-202510", "output_type": "paper", "title": "Self-supervised anomaly detection for HPLC traces in a cloud laboratory", "date": "2025-10-08", "venue": "Digital Discovery", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "reported", "source_ids": [ "rsc-ecl-hplc-anomaly-202510", "official-emerald-cloud-lab" ], "entity_ids": [ "emerald-cloud-lab" ], "entity_names": [ "Emerald Cloud Lab" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "strateos-cloud-lab-platform-profile-202605", "output_type": "platform_report", "title": "Cloud laboratory automation platform public technical profile", "date": "2026-05-11", "venue": "Strateos", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-strateos" ], "entity_ids": [ "strateos" ], "entity_names": [ "Strateos" ] }, { "doi": "10.1126/science.abc2986", "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "chemify-chemputer-universal-synthesis-202012", "output_type": "paper", "title": "A universal system for digitization and automatic execution of the chemical synthesis literature", "date": "2020-12-04", "venue": "Science", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "reported", "source_ids": [ "nature-chemputer-202012", "official-chemify-technology" ], "entity_ids": [ "chemify" ], "entity_names": [ "Chemify" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "chemify-chemputer-blueprints-202411", "output_type": "paper", "title": "Automatic conversion of chemical synthesis procedures into digital executable blueprints", "date": "2024-11-14", "venue": "Nature Communications", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "reported", "source_ids": [ "nature-comm-chemputer-blueprints-202411", "official-chemify-technology" ], "entity_ids": [ "chemify" ], "entity_names": [ "Chemify" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "chemify-llm-chemputation-202604", "output_type": "paper", "title": "Chemputation with large language models", "date": "2026-04-14", "venue": "Communications Chemistry", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "reported", "source_ids": [ "comms-chem-chemputation-llm-202604", "official-chemify-technology" ], "entity_ids": [ "chemify" ], "entity_names": [ "Chemify" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "atinary-semopt-constraints-202205", "output_type": "preprint", "title": "Bayesian optimization for self-driving labs with unknown constraints", "date": "2022-05-09", "venue": "ChemRxiv", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "chemrxiv-semopt-202205", "official-atinary-semopt" ], "entity_ids": [ "atinary" ], "entity_names": [ "Atinary" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "pnnl-autonomous-science-profile-202605", "output_type": "platform_report", "title": "Autonomous science and self-driving laboratories program profile", "date": "2026-05-11", "venue": "PNNL", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-pnnl-autonomous-science", "official-pnnl" ], "entity_ids": [ "pnnl" ], "entity_names": [ "PNNL" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "pnnl-energy-storage-hte-202605", "output_type": "platform_report", "title": "High-throughput experimentation for energy storage materials", "date": "2026-05-11", "venue": "PNNL", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-pnnl-energy-storage-hte", "official-pnnl" ], "entity_ids": [ "pnnl" ], "entity_names": [ "PNNL" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "acceleration-consortium-sdl-acceleration-202504", "output_type": "paper", "title": "Acceleration of scientific discovery by self-driving laboratories", "date": "2025-04-04", "venue": "Nature Communications", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "reported", "source_ids": [ "nature-comm-self-driving-labs-202504", "official-acceleration-consortium" ], "entity_ids": [ "acceleration-consortium" ], "entity_names": [ "Acceleration Consortium" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "acceleration-consortium-rise-sdl-202301", "output_type": "review", "title": "The rise of self-driving labs in chemical and materials sciences", "date": "2023-01-12", "venue": "Nature Synthesis", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "reported", "source_ids": [ "nature-synthesis-rise-sdl-202301", "official-acceleration-consortium" ], "entity_ids": [ "acceleration-consortium" ], "entity_names": [ "Acceleration Consortium" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "radical-ai-platform-profile-202605", "output_type": "platform_report", "title": "Radical AI autonomous science platform public technical profile", "date": "2026-05-11", "venue": "Radical AI", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-radical-ai-platform", "official-radical-ai" ], "entity_ids": [ "radical-ai" ], "entity_names": [ "Radical AI" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "xaira-platform-profile-202605", "output_type": "platform_report", "title": "Integrated AI drug discovery platform public technical profile", "date": "2026-05-11", "venue": "Xaira Therapeutics", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-xaira-platform", "official-xaira-therapeutics" ], "entity_ids": [ "xaira-therapeutics" ], "entity_names": [ "Xaira Therapeutics" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "helical-virtual-lab-profile-202605", "output_type": "platform_report", "title": "Virtual AI lab for reproducible in-silico drug discovery", "date": "2026-05-11", "venue": "Helical", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-helix-virtual-lab", "official-helical" ], "entity_ids": [ "helical" ], "entity_names": [ "Helical" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "fathom-physics-ai-platform-profile-202605", "output_type": "platform_report", "title": "Physics and AI-enabled small-molecule design platform profile", "date": "2026-05-11", "venue": "Fathom Therapeutics", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-fathom-platform", "official-fathom-therapeutics" ], "entity_ids": [ "fathom-therapeutics" ], "entity_names": [ "Fathom Therapeutics" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "tenx-science-protein-characterization-profile-202605", "output_type": "platform_report", "title": "AI-native protein characterization platform profile", "date": "2026-05-11", "venue": "10x Science", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-10x-science-platform", "yc-10x-science-profile-202604", "prnewswire-10x-science-seed-202604" ], "entity_ids": [ "10x-science" ], "entity_names": [ "10x Science" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "kebotix-platform-profile-202605", "output_type": "platform_report", "title": "AI and robotics materials discovery platform profile", "date": "2026-05-11", "venue": "Kebotix", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-kebotix" ], "entity_ids": [ "kebotix" ], "entity_names": [ "Kebotix" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "citrine-citrination-platform-201609", "output_type": "paper", "title": "The Citrination platform: a materials data facility", "date": "2016-09-27", "venue": "JOM", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "reported", "source_ids": [ "citrine-materials-data-facility-2016", "official-citrine-informatics" ], "entity_ids": [ "citrine-informatics" ], "entity_names": [ "Citrine Informatics" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "artificial-lab-automation-profile-202605", "output_type": "platform_report", "title": "Artificial lab automation execution platform profile", "date": "2026-05-11", "venue": "Artificial", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-artificial" ], "entity_ids": [ "artificial" ], "entity_names": [ "Artificial" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "synthace-experiment-platform-profile-202605", "output_type": "platform_report", "title": "Experiment design and automation platform profile", "date": "2026-05-11", "venue": "Synthace", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-synthace-how-it-works", "official-synthace" ], "entity_ids": [ "synthace" ], "entity_names": [ "Synthace" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "tetrascience-tetra-data-platform-202605", "output_type": "platform_report", "title": "Tetra Data Platform and Tetra OS public technical profile", "date": "2026-05-11", "venue": "TetraScience", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-tetrascience-tetra-data", "official-tetrascience-tetra-os", "official-tetrascience" ], "entity_ids": [ "tetrascience" ], "entity_names": [ "TetraScience" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "riffyn-x-enterprise-recipe-profile-202605", "output_type": "platform_report", "title": "Riffyn X enterprise recipe and process data platform profile", "date": "2026-05-11", "venue": "Siemens Software", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "siemens-riffyn-x" ], "entity_ids": [ "riffyn" ], "entity_names": [ "Riffyn" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "opentrons-flex-docs-202605", "output_type": "platform_report", "title": "Opentrons Flex robot public technical documentation", "date": "2026-05-11", "venue": "Opentrons", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "docs-opentrons-flex", "official-opentrons-flex" ], "entity_ids": [ "opentrons" ], "entity_names": [ "Opentrons" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "tecan-fluentcontrol-profile-202605", "output_type": "platform_report", "title": "Fluent and FluentControl lab automation technical profile", "date": "2026-05-11", "venue": "Tecan", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-tecan-fluent", "official-tecan-fluentcontrol", "official-tecan-history" ], "entity_ids": [ "tecan" ], "entity_names": [ "Tecan" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "highres-cellario-profile-202605", "output_type": "platform_report", "title": "Cellario laboratory automation scheduler public technical profile", "date": "2026-05-11", "venue": "HighRes Biosolutions", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-highres-cellario", "official-highres" ], "entity_ids": [ "highres-biosolutions" ], "entity_names": [ "HighRes Biosolutions" ] }, { "doi": "10.1126/science.adu7875", "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "cellarity-science-framework-202510", "output_type": "paper", "title": "A framework for the discovery of cell state-correcting medicines", "date": "2025-10-03", "venue": "Science", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "reported", "source_ids": [ "science-cellarity-framework-202510", "official-cellarity-science-framework-202510" ], "entity_ids": [ "cellarity" ], "entity_names": [ "Cellarity" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": true, "model_available": null, "last_verified": "2026-05-11", "output_id": "cellarity-open-single-cell-dataset-202208", "output_type": "dataset_release", "title": "Open single-cell dataset for cell-based drug discovery", "date": "2022-08-23", "venue": "Cellarity", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "cellarity-open-single-cell-dataset-202208", "official-cellarity-about" ], "entity_ids": [ "cellarity" ], "entity_names": [ "Cellarity" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "abcellera-lilly-covid-antibody-202003", "output_type": "collaboration_report", "title": "Rapid pandemic antibody discovery collaboration with Lilly", "date": "2020-03-12", "venue": "AbCellera", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-abcellera-lilly-covid-202003", "official-abcellera-platform" ], "entity_ids": [ "abcellera" ], "entity_names": [ "AbCellera" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "iambic-iam1363-first-patient-202403", "output_type": "clinical_program_report", "title": "IAM1363 Phase 1 first-patient-dosed clinical program report", "date": "2024-03-28", "venue": "Iambic Therapeutics", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-iambic-iam1363-first-patient-202403", "clinicaltrials-nct06253871" ], "entity_ids": [ "iambic-therapeutics" ], "entity_names": [ "Iambic Therapeutics" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "insilico-rentosertib-natmed-202506", "output_type": "paper", "title": "A generative AI-discovered TNIK inhibitor for idiopathic pulmonary fibrosis", "date": "2025-06-01", "venue": "Nature Medicine", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "reported", "source_ids": [ "nature-med-insilico-rentosertib-202506", "official-insilico-rentosertib-natmed-202506", "clinicaltrials-nct05938920" ], "entity_ids": [ "insilico-medicine" ], "entity_names": [ "Insilico Medicine" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "ginkgo-bioworks-automation-platform-202605", "output_type": "platform_report", "title": "Ginkgo Automation platform public technical profile", "date": "2026-05-11", "venue": "Ginkgo Bioworks", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-ginkgo-automation-platform", "official-ginkgo-bioworks" ], "entity_ids": [ "ginkgo-bioworks" ], "entity_names": [ "Ginkgo Bioworks" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "ginkgo-biological-engineering-language-201909", "output_type": "paper", "title": "A programming language for biological engineering", "date": "2019-09-23", "venue": "Nature Biotechnology", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "reported", "source_ids": [ "nature-ginkgo-operating-system-201909", "official-ginkgo-bioworks" ], "entity_ids": [ "ginkgo-bioworks" ], "entity_names": [ "Ginkgo Bioworks" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "ami-labs-platform-profile-202605", "output_type": "platform_report", "title": "AI-native experimental workflow platform profile", "date": "2026-05-11", "venue": "AMI Labs", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-ami-labs-platform", "official-ami-labs" ], "entity_ids": [ "ami-labs" ], "entity_names": [ "AMI Labs" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "chemlex-platform-profile-202605", "output_type": "platform_report", "title": "Chemistry reasoning and lab workflow platform profile", "date": "2026-05-11", "venue": "ChemLex", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-chemlex-platform", "official-chemlex" ], "entity_ids": [ "chemlex" ], "entity_names": [ "ChemLex" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "genesis-molecular-ai-platform-profile-202605", "output_type": "platform_report", "title": "Genesis molecular AI platform public technical profile", "date": "2026-05-11", "venue": "Genesis Molecular AI", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-genesis-platform", "official-genesis-molecular-ai" ], "entity_ids": [ "genesis-molecular-ai" ], "entity_names": [ "Genesis Molecular AI" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "brookhaven-ai-nanostructures-202301", "output_type": "paper", "title": "AI-guided autonomous nanostructure discovery from X-ray scattering", "date": "2023-01-11", "venue": "Science Advances", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "reported", "source_ids": [ "science-advances-bnl-nanostructures-202301", "official-bnl-ai-nanostructures-202301", "official-bnl-nsls2-dssi" ], "entity_ids": [ "brookhaven-national-laboratory" ], "entity_names": [ "Brookhaven National Laboratory" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "materials-genome-autonomous-experimentation-202605", "output_type": "program_report", "title": "Autonomous Experimentation for Materials R&D report", "date": "2026-05-11", "venue": "Materials Genome Initiative", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-mgi-autonomous-experimentation-report", "official-mgi-about" ], "entity_ids": [ "materials-genome-initiative" ], "entity_names": [ "Materials Genome Initiative" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "nist-composable-lab-ecosystem-202605", "output_type": "paper", "title": "A composable laboratory ecosystem for autonomous experimentation", "date": "2026-05-07", "venue": "NIST", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "reported", "source_ids": [ "nist-composable-lab-ecosystem-202605", "official-nist-autonomous-labs" ], "entity_ids": [ "nist" ], "entity_names": [ "NIST" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "nrel-text-to-test-202410", "output_type": "paper", "title": "From text to test: AI-generated control software for materials science instruments", "date": "2024-10-23", "venue": "Digital Discovery", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "reported", "source_ids": [ "rsc-text-to-test-202410", "official-nrel-autonomous-experimentation" ], "entity_ids": [ "nrel" ], "entity_names": [ "NREL" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": false, "last_verified": "2026-05-11", "output_id": "openai-gpt-rosalind-202604", "output_type": "model_release", "title": "GPT-Rosalind life-sciences model release", "date": "2026-04-16", "venue": "OpenAI", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "openai-gpt-rosalind-202604", "fiercebiotech-gpt-rosalind-202604" ], "entity_ids": [ "openai" ], "entity_names": [ "OpenAI" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": true, "model_available": null, "last_verified": "2026-05-11", "output_id": "openai-genebench-report-202604", "output_type": "benchmark_report", "title": "GeneBench benchmark report for genetics and quantitative biology agents", "date": "2026-04-23", "venue": "OpenAI", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "openai-genebench-202604", "biorxiv-genebench-202604" ], "entity_ids": [ "openai" ], "entity_names": [ "OpenAI" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "causaly-knowledge-graph-platform-202605", "output_type": "platform_report", "title": "Biomedical knowledge graph and Bio Graph platform profile", "date": "2026-05-11", "venue": "Causaly", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-causaly-technology", "official-causaly-knowledge-graph", "official-causaly-bio-graph" ], "entity_ids": [ "causaly" ], "entity_names": [ "Causaly" ] }, { "doi": "10.1186/s13321-023-00742-8", "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "iktos-synthetic-accessibility-generative-design-202309", "output_type": "paper", "title": "Integrating synthetic accessibility with AI-based generative drug design", "date": "2023-09-19", "venue": "Journal of Cheminformatics", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "reported", "source_ids": [ "jcheminf-iktos-synthetic-accessibility-202309", "official-iktos-makya" ], "entity_ids": [ "iktos" ], "entity_names": [ "Iktos" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-11", "output_id": "iktos-growing-linking-optimizers-202509", "output_type": "paper", "title": "Growing and Linking Optimizers: synthesis-driven molecule design", "date": "2025-09-25", "venue": "Iktos / Briefings in Bioinformatics", "peer_reviewed": true, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "official-iktos-publications-go-lo-202509", "official-iktos-makya" ], "entity_ids": [ "iktos" ], "entity_names": [ "Iktos" ] }, { "doi": "10.64898/2026.02.05.703998", "arxiv_id": null, "biorxiv_id": "2026.02.05.703998", "pubmed_id": null, "license": "CC BY 4.0", "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-13", "output_id": "openai-ginkgo-gpt5-cfps-autonomous-lab-2026", "output_type": "preprint", "title": "Using a GPT-5-driven autonomous lab to optimize the cost and titer of cell-free protein synthesis", "date": "2026-02-05", "venue": "bioRxiv / OpenAI / Ginkgo Bioworks", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "openai-ginkgo-cfps-autonomous-lab-202602", "ginkgo-openai-cfps-202602", "openai-ginkgo-cfps-preprint-202602", "biorxiv-openai-ginkgo-cfps-202602" ], "entity_ids": [ "openai", "ginkgo-bioworks" ], "entity_names": [ "OpenAI", "Ginkgo Bioworks" ] }, { "doi": "10.1101/2025.07.05.663018", "arxiv_id": null, "biorxiv_id": "2025.07.05.663018", "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-13", "output_id": "chai2-zero-shot-antibody-design-202507", "output_type": "preprint", "title": "Zero-shot antibody design in a 24-well plate", "date": "2025-07-06", "venue": "bioRxiv / Chai Discovery", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "businesswire-chai2-202506", "biorxiv-chai2-zero-shot-202507" ], "entity_ids": [ "chai-discovery" ], "entity_names": [ "Chai Discovery" ] }, { "doi": "10.1038/s44160-026-01053-0", "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": "CC BY 4.0", "code_available": true, "dataset_available": null, "model_available": null, "last_verified": "2026-05-13", "output_id": "robochem-flex-self-driving-lab-202604", "output_type": "paper", "title": "A flexible and affordable self-driving laboratory for automated reaction optimization", "date": "2026-04-13", "venue": "Nature Synthesis", "peer_reviewed": true, "open_access": true, "source_tier": "A", "claim_status": "confirmed", "source_ids": [ "nature-robochem-flex-202604" ], "entity_ids": [ "robochem-flex" ], "entity_names": [ "RoboChem-Flex" ] }, { "doi": null, "arxiv_id": null, "biorxiv_id": null, "pubmed_id": null, "license": null, "code_available": null, "dataset_available": null, "model_available": null, "last_verified": "2026-05-13", "output_id": "generare-microbial-chemistry-platform-202604", "output_type": "platform_profile", "title": "Generare microbial-genome platform for silent chemistry and novel small-molecule data generation", "date": "2026-04-02", "venue": "Generare / Alven", "peer_reviewed": false, "open_access": true, "source_tier": "B", "claim_status": "reported", "source_ids": [ "alven-generare-series-a-202604", "biologydigital-generare-series-a-202604" ], "entity_ids": [ "generare" ], "entity_names": [ "Generare" ] } ], "benchmark-results": [ { "result_id": "genebench-gpt-55-pro-overall-202604", "benchmark_id": "genebench", "benchmark_name": "GeneBench", "task": "multi_stage_genomics_quantitative_biology", "domain": "genomics", "evaluated_system_id": "openai", "evaluated_system_name": "GPT-5.5 Pro", "system_type": "frontier_model", "model_or_version": "GPT-5.5 Pro, reported April 2026", "metric": "overall_pass_rate", "score": 0.332, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "Gemini 3.1 Pro external baseline", "baseline_score": 0.112, "evaluation_date": "2026-04", "benchmark_version": "v1, 103 problems", "dataset_split": "full benchmark", "evaluator_type": "self_reported", "evaluator_name": "OpenAI", "reproducibility_status": "not_independently_reproduced", "claim_status": "reported", "source_tier": "B", "source_ids": [ "openai-genebench-202604", "biorxiv-genebench-202604" ], "notes": "OpenAI GeneBench report table; score is unweighted mean per-problem pass rate." }, { "result_id": "genebench-gpt-55-xhigh-overall-202604", "benchmark_id": "genebench", "benchmark_name": "GeneBench", "task": "multi_stage_genomics_quantitative_biology", "domain": "genomics", "evaluated_system_id": "openai", "evaluated_system_name": "GPT-5.5 xhigh", "system_type": "frontier_model", "model_or_version": "GPT-5.5 xhigh", "metric": "overall_pass_rate", "score": 0.25, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "Gemini 3.1 Pro external baseline", "baseline_score": 0.112, "evaluation_date": "2026-04", "benchmark_version": "v1, 103 problems", "dataset_split": "full benchmark", "evaluator_type": "self_reported", "evaluator_name": "OpenAI", "reproducibility_status": "not_independently_reproduced", "claim_status": "reported", "source_tier": "B", "source_ids": [ "openai-genebench-202604", "biorxiv-genebench-202604" ], "notes": "OpenAI GeneBench report table; score is unweighted mean per-problem pass rate." }, { "result_id": "genebench-gpt-54-pro-overall-202604", "benchmark_id": "genebench", "benchmark_name": "GeneBench", "task": "multi_stage_genomics_quantitative_biology", "domain": "genomics", "evaluated_system_id": "openai", "evaluated_system_name": "GPT-5.4 Pro", "system_type": "frontier_model", "model_or_version": "GPT-5.4 Pro", "metric": "overall_pass_rate", "score": 0.256, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "Gemini 3.1 Pro external baseline", "baseline_score": 0.112, "evaluation_date": "2026-04", "benchmark_version": "v1, 103 problems", "dataset_split": "full benchmark", "evaluator_type": "self_reported", "evaluator_name": "OpenAI", "reproducibility_status": "not_independently_reproduced", "claim_status": "reported", "source_tier": "B", "source_ids": [ "openai-genebench-202604", "biorxiv-genebench-202604" ], "notes": "OpenAI GeneBench report table; score is unweighted mean per-problem pass rate." }, { "result_id": "genebench-gpt-54-xhigh-overall-202604", "benchmark_id": "genebench", "benchmark_name": "GeneBench", "task": "multi_stage_genomics_quantitative_biology", "domain": "genomics", "evaluated_system_id": "openai", "evaluated_system_name": "GPT-5.4 xhigh", "system_type": "frontier_model", "model_or_version": "GPT-5.4 xhigh", "metric": "overall_pass_rate", "score": 0.19, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "GPT-5.2 xhigh", "baseline_score": 0.094, "evaluation_date": "2026-04", "benchmark_version": "v1, 103 problems", "dataset_split": "full benchmark", "evaluator_type": "self_reported", "evaluator_name": "OpenAI", "reproducibility_status": "not_independently_reproduced", "claim_status": "reported", "source_tier": "B", "source_ids": [ "openai-genebench-202604", "biorxiv-genebench-202604" ], "notes": "OpenAI GeneBench report table; score is unweighted mean per-problem pass rate." }, { "result_id": "genebench-gpt-52-pro-overall-202604", "benchmark_id": "genebench", "benchmark_name": "GeneBench", "task": "multi_stage_genomics_quantitative_biology", "domain": "genomics", "evaluated_system_id": "openai", "evaluated_system_name": "GPT-5.2 Pro", "system_type": "frontier_model", "model_or_version": "GPT-5.2 Pro", "metric": "overall_pass_rate", "score": 0.108, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "Gemini 3.1 Pro external baseline", "baseline_score": 0.112, "evaluation_date": "2026-04", "benchmark_version": "v1, 103 problems", "dataset_split": "full benchmark", "evaluator_type": "self_reported", "evaluator_name": "OpenAI", "reproducibility_status": "not_independently_reproduced", "claim_status": "reported", "source_tier": "B", "source_ids": [ "openai-genebench-202604", "biorxiv-genebench-202604" ], "notes": "OpenAI GeneBench report table; this OpenAI model is slightly below the named external baseline." }, { "result_id": "genebench-gemini-31-pro-overall-202604", "benchmark_id": "genebench", "benchmark_name": "GeneBench", "task": "multi_stage_genomics_quantitative_biology", "domain": "genomics", "evaluated_system_id": "google-deepmind", "evaluated_system_name": "Gemini 3.1 Pro", "system_type": "frontier_model", "model_or_version": "Gemini 3.1 Pro high", "metric": "overall_pass_rate", "score": 0.112, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "GPT-5.2 xhigh", "baseline_score": 0.094, "evaluation_date": "2026-04", "benchmark_version": "v1, 103 problems", "dataset_split": "full benchmark", "evaluator_type": "self_reported", "evaluator_name": "OpenAI", "reproducibility_status": "not_independently_reproduced", "claim_status": "reported", "source_tier": "B", "source_ids": [ "openai-genebench-202604", "biorxiv-genebench-202604" ], "notes": "External-model baseline reported in OpenAI GeneBench table." }, { "result_id": "genebench-kimi-k26-overall-202604", "benchmark_id": "genebench", "benchmark_name": "GeneBench", "task": "multi_stage_genomics_quantitative_biology", "domain": "genomics", "evaluated_system_id": null, "evaluated_system_name": "Kimi K2.6", "system_type": "frontier_model", "model_or_version": "Kimi K2.6", "metric": "overall_pass_rate", "score": 0.074, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "GLM 5.1", "baseline_score": 0.042, "evaluation_date": "2026-04", "benchmark_version": "v1, 103 problems", "dataset_split": "full benchmark", "evaluator_type": "self_reported", "evaluator_name": "OpenAI", "reproducibility_status": "not_independently_reproduced", "claim_status": "reported", "source_tier": "B", "source_ids": [ "openai-genebench-202604", "biorxiv-genebench-202604" ], "notes": "External-model comparison row from the GeneBench report table." }, { "result_id": "bixbench-gpt-rosalind-pass1-202604", "benchmark_id": "bixbench", "benchmark_name": "BixBench", "task": "bioinformatics_data_analysis", "domain": "bioinformatics", "evaluated_system_id": "openai", "evaluated_system_name": "GPT-Rosalind", "system_type": "frontier_model", "model_or_version": "GPT-Rosalind research preview, reported April 2026", "metric": "pass_at_1", "score": 0.751, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "GPT-5.4", "baseline_score": 0.732, "evaluation_date": "2026-04-16", "benchmark_version": "BixBench public benchmark, OpenAI launch chart", "dataset_split": "published-score comparison", "evaluator_type": "self_reported", "evaluator_name": "OpenAI", "reproducibility_status": "not_independently_reproduced", "claim_status": "reported", "source_tier": "B", "source_ids": [ "openai-gpt-rosalind-202604", "the-decoder-gpt-rosalind-bixbench-202604" ], "notes": "Numbers are from OpenAI's launch chart and transcribed in The Decoder coverage: GPT-Rosalind 0.751 Pass@1 versus GPT-5.4 0.732." }, { "result_id": "labbench-claude35-tableqa-202407", "benchmark_id": "lab-bench", "benchmark_name": "LAB-Bench", "task": "table_qa", "domain": "biology", "evaluated_system_id": null, "evaluated_system_name": "Claude 3.5 Sonnet", "system_type": "frontier_model", "model_or_version": "claude-3-5-sonnet-20240620", "metric": "accuracy", "score": 0.83, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "GPT-4o", "baseline_score": 0.71, "evaluation_date": "2024-07", "benchmark_version": "LAB-Bench v1", "dataset_split": "public and private splits", "evaluator_type": "independent", "evaluator_name": "LAB-Bench authors", "reproducibility_status": "unknown", "claim_status": "reported", "source_tier": "B", "source_ids": [ "arxiv-labbench-202407" ], "notes": "Table 2 accuracy averaged over three model runs across the full LAB-Bench dataset." }, { "result_id": "labbench-gpt4o-tableqa-202407", "benchmark_id": "lab-bench", "benchmark_name": "LAB-Bench", "task": "table_qa", "domain": "biology", "evaluated_system_id": "openai", "evaluated_system_name": "GPT-4o", "system_type": "frontier_model", "model_or_version": "gpt-4o", "metric": "accuracy", "score": 0.71, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "Human expert baseline", "baseline_score": 0.84, "evaluation_date": "2024-07", "benchmark_version": "LAB-Bench v1", "dataset_split": "public and private splits", "evaluator_type": "independent", "evaluator_name": "LAB-Bench authors", "reproducibility_status": "unknown", "claim_status": "reported", "source_tier": "B", "source_ids": [ "arxiv-labbench-202407" ], "notes": "Table 2 accuracy averaged over three model runs across the full LAB-Bench dataset." }, { "result_id": "labbench-claude35-protocolqa-202407", "benchmark_id": "lab-bench", "benchmark_name": "LAB-Bench", "task": "protocol_qa", "domain": "biology", "evaluated_system_id": null, "evaluated_system_name": "Claude 3.5 Sonnet", "system_type": "frontier_model", "model_or_version": "claude-3-5-sonnet-20240620", "metric": "accuracy", "score": 0.48, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "GPT-4o", "baseline_score": 0.53, "evaluation_date": "2024-07", "benchmark_version": "LAB-Bench v1", "dataset_split": "public and private splits", "evaluator_type": "independent", "evaluator_name": "LAB-Bench authors", "reproducibility_status": "unknown", "claim_status": "reported", "source_tier": "B", "source_ids": [ "arxiv-labbench-202407" ], "notes": "Table 2 accuracy averaged over three model runs across the full LAB-Bench dataset." }, { "result_id": "labbench-gpt4o-protocolqa-202407", "benchmark_id": "lab-bench", "benchmark_name": "LAB-Bench", "task": "protocol_qa", "domain": "biology", "evaluated_system_id": "openai", "evaluated_system_name": "GPT-4o", "system_type": "frontier_model", "model_or_version": "gpt-4o", "metric": "accuracy", "score": 0.53, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "Human expert baseline", "baseline_score": 0.79, "evaluation_date": "2024-07", "benchmark_version": "LAB-Bench v1", "dataset_split": "public and private splits", "evaluator_type": "independent", "evaluator_name": "LAB-Bench authors", "reproducibility_status": "unknown", "claim_status": "reported", "source_tier": "B", "source_ids": [ "arxiv-labbench-202407" ], "notes": "Table 2 accuracy averaged over three model runs across the full LAB-Bench dataset." }, { "result_id": "labbench-claude35-litqa2-202407", "benchmark_id": "lab-bench", "benchmark_name": "LAB-Bench", "task": "literature_qa", "domain": "biology", "evaluated_system_id": null, "evaluated_system_name": "Claude 3.5 Sonnet", "system_type": "frontier_model", "model_or_version": "claude-3-5-sonnet-20240620", "metric": "accuracy", "score": 0.06, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "Human expert baseline", "baseline_score": 0.7, "evaluation_date": "2024-07", "benchmark_version": "LAB-Bench v1", "dataset_split": "public and private splits", "evaluator_type": "independent", "evaluator_name": "LAB-Bench authors", "reproducibility_status": "unknown", "claim_status": "reported", "source_tier": "B", "source_ids": [ "arxiv-labbench-202407" ], "notes": "Table 2 LitQA2 accuracy averaged over three model runs." }, { "result_id": "labbench-gpt4o-litqa2-202407", "benchmark_id": "lab-bench", "benchmark_name": "LAB-Bench", "task": "literature_qa", "domain": "biology", "evaluated_system_id": "openai", "evaluated_system_name": "GPT-4o", "system_type": "frontier_model", "model_or_version": "gpt-4o", "metric": "accuracy", "score": 0.27, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "Human expert baseline", "baseline_score": 0.7, "evaluation_date": "2024-07", "benchmark_version": "LAB-Bench v1", "dataset_split": "public and private splits", "evaluator_type": "independent", "evaluator_name": "LAB-Bench authors", "reproducibility_status": "unknown", "claim_status": "reported", "source_tier": "B", "source_ids": [ "arxiv-labbench-202407" ], "notes": "Table 2 LitQA2 accuracy averaged over three model runs." }, { "result_id": "labbench-openresponse-claude35-protocolqa-202407", "benchmark_id": "lab-bench", "benchmark_name": "LAB-Bench", "task": "open_response_protocol_qa", "domain": "biology", "evaluated_system_id": null, "evaluated_system_name": "Claude 3.5 Sonnet", "system_type": "frontier_model", "model_or_version": "claude-3-5-sonnet-20240620", "metric": "open_response_accuracy", "score": 0.3, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "GPT-4o", "baseline_score": 0.2, "evaluation_date": "2024-07", "benchmark_version": "LAB-Bench v1 open-response subset", "dataset_split": "open-response subset", "evaluator_type": "independent", "evaluator_name": "LAB-Bench expert biologist graders", "reproducibility_status": "unknown", "claim_status": "reported", "source_tier": "B", "source_ids": [ "arxiv-labbench-202407" ], "notes": "Supplemental Table 5 open-response subset, graded by an expert biologist author and reviewed by a second." }, { "result_id": "bixbench-claude35-open-answer-202503", "benchmark_id": "bixbench", "benchmark_name": "BixBench", "task": "open_answer_bioinformatics", "domain": "bioinformatics", "evaluated_system_id": null, "evaluated_system_name": "Claude 3.5 Sonnet", "system_type": "research_agent", "model_or_version": "Claude 3.5 Sonnet agent configuration", "metric": "accuracy", "score": 0.21, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "GPT-4o agent baseline", "baseline_score": 0.15, "evaluation_date": "2025-03", "benchmark_version": "BixBench v1", "dataset_split": "open-answer regime", "evaluator_type": "self_reported", "evaluator_name": "FutureHouse", "reproducibility_status": "not_independently_reproduced", "claim_status": "reported", "source_tier": "B", "source_ids": [ "arxiv-bixbench-202503", "official-futurehouse-bixbench-202503" ], "notes": "BixBench paper reports Claude 3.5 Sonnet bested GPT-4o with 21% versus 15% open-answer accuracy." }, { "result_id": "paperqa2-litqa2-precision-202409", "benchmark_id": "litqa2-paperqa", "benchmark_name": "LitQA2 / PaperQA2", "task": "scientific_literature_qa", "domain": "scientific literature", "evaluated_system_id": "futurehouse", "evaluated_system_name": "PaperQA2", "system_type": "research_agent", "model_or_version": "PaperQA2 reported 2024-09", "metric": "precision", "score": 0.852, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "Human PhD annotators", "baseline_score": 0.738, "evaluation_date": "2024-09", "benchmark_version": "LitQA2, 248 questions", "dataset_split": "full benchmark", "evaluator_type": "self_reported", "evaluator_name": "PaperQA2 authors", "reproducibility_status": "not_independently_reproduced", "claim_status": "reported", "source_tier": "B", "source_ids": [ "arxiv-paperqa2-202409" ], "notes": "Mean precision over three PaperQA2 runs compared with nine human annotators." }, { "result_id": "paperqa2-litqa2-accuracy-202409", "benchmark_id": "litqa2-paperqa", "benchmark_name": "LitQA2 / PaperQA2", "task": "scientific_literature_qa", "domain": "scientific literature", "evaluated_system_id": "futurehouse", "evaluated_system_name": "PaperQA2", "system_type": "research_agent", "model_or_version": "PaperQA2 reported 2024-09", "metric": "accuracy", "score": 0.66, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "Human PhD annotators", "baseline_score": 0.677, "evaluation_date": "2024-09", "benchmark_version": "LitQA2, 248 questions", "dataset_split": "full benchmark", "evaluator_type": "self_reported", "evaluator_name": "PaperQA2 authors", "reproducibility_status": "not_independently_reproduced", "claim_status": "reported", "source_tier": "B", "source_ids": [ "arxiv-paperqa2-202409" ], "notes": "Mean accuracy over three PaperQA2 runs; paper reports no statistically significant difference from human accuracy." }, { "result_id": "paperqa2-gpt4turbo-old147-accuracy-202409", "benchmark_id": "litqa2-paperqa", "benchmark_name": "LitQA2 / PaperQA2", "task": "scientific_literature_qa_model_ablation", "domain": "scientific literature", "evaluated_system_id": "openai", "evaluated_system_name": "GPT-4 Turbo in PaperQA2 RCS", "system_type": "research_agent_component", "model_or_version": "GPT-4 Turbo RCS ablation", "metric": "accuracy", "score": 0.644, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "Prior PaperQA configuration", "baseline_score": 0.367, "evaluation_date": "2024-09", "benchmark_version": "LitQA2 Table 2", "dataset_split": "old 147 question set", "evaluator_type": "self_reported", "evaluator_name": "PaperQA2 authors", "reproducibility_status": "not_independently_reproduced", "claim_status": "reported", "source_tier": "B", "source_ids": [ "arxiv-paperqa2-202409" ], "notes": "Table 2 model-choice ablation; baseline is prior published PaperQA configuration reported in the same paper." }, { "result_id": "paperqa2-gpt4turbo-new101-accuracy-202409", "benchmark_id": "litqa2-paperqa", "benchmark_name": "LitQA2 / PaperQA2", "task": "scientific_literature_qa_model_ablation", "domain": "scientific literature", "evaluated_system_id": "openai", "evaluated_system_name": "GPT-4 Turbo in PaperQA2 RCS", "system_type": "research_agent_component", "model_or_version": "GPT-4 Turbo RCS ablation", "metric": "accuracy", "score": 0.637, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "Prior PaperQA configuration", "baseline_score": 0.367, "evaluation_date": "2024-09", "benchmark_version": "LitQA2 Table 2", "dataset_split": "new 101 question set", "evaluator_type": "self_reported", "evaluator_name": "PaperQA2 authors", "reproducibility_status": "not_independently_reproduced", "claim_status": "reported", "source_tier": "B", "source_ids": [ "arxiv-paperqa2-202409" ], "notes": "Table 2 model-choice ablation; baseline is prior published PaperQA configuration reported in the same paper." }, { "result_id": "foldbench-af3-ppi-dockq-auc-202512", "benchmark_id": "foldbench", "benchmark_name": "FoldBench", "task": "protein_protein_interface_prediction", "domain": "biology", "evaluated_system_id": "google-deepmind", "evaluated_system_name": "AlphaFold 3", "system_type": "structure_prediction_model", "model_or_version": "AlphaFold 3 FoldBench evaluation", "metric": "dockq_auc", "score": 0.59, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "Boltz-1", "baseline_score": 0.53, "evaluation_date": "2025-12", "benchmark_version": "FoldBench n=1522 assemblies", "dataset_split": "279 low-homology protein-protein interfaces", "evaluator_type": "independent", "evaluator_name": "FoldBench authors", "reproducibility_status": "independently_reproduced", "claim_status": "reported", "source_tier": "A", "source_ids": [ "nature-foldbench-2025" ], "notes": "FoldBench reports cumulative DockQ-score AUC for 279 protein-protein targets." }, { "result_id": "foldbench-af3-ppi-success-202512", "benchmark_id": "foldbench", "benchmark_name": "FoldBench", "task": "protein_protein_interface_prediction", "domain": "biology", "evaluated_system_id": "google-deepmind", "evaluated_system_name": "AlphaFold 3", "system_type": "structure_prediction_model", "model_or_version": "AlphaFold 3 FoldBench evaluation", "metric": "dockq_success_rate", "score": 0.729, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "Chai-1", "baseline_score": 0.685, "evaluation_date": "2025-12", "benchmark_version": "FoldBench n=1522 assemblies", "dataset_split": "279 low-homology protein-protein interfaces", "evaluator_type": "independent", "evaluator_name": "FoldBench authors", "reproducibility_status": "independently_reproduced", "claim_status": "reported", "source_tier": "A", "source_ids": [ "nature-foldbench-2025" ], "notes": "DockQ success threshold >= 0.23 on protein-protein interfaces." }, { "result_id": "foldbench-af3-antibody-antigen-auc-202512", "benchmark_id": "foldbench", "benchmark_name": "FoldBench", "task": "antibody_antigen_interface_prediction", "domain": "biology", "evaluated_system_id": "google-deepmind", "evaluated_system_name": "AlphaFold 3", "system_type": "structure_prediction_model", "model_or_version": "AlphaFold 3 FoldBench evaluation", "metric": "dockq_auc", "score": 0.36, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "Protenix", "baseline_score": 0.23, "evaluation_date": "2025-12", "benchmark_version": "FoldBench n=1522 assemblies", "dataset_split": "172 antibody-antigen pairs", "evaluator_type": "independent", "evaluator_name": "FoldBench authors", "reproducibility_status": "independently_reproduced", "claim_status": "reported", "source_tier": "A", "source_ids": [ "nature-foldbench-2025" ], "notes": "FoldBench reports antibody-antigen cumulative DockQ-score AUC." }, { "result_id": "foldbench-af3-dna-monomer-lddt-202512", "benchmark_id": "foldbench", "benchmark_name": "FoldBench", "task": "dna_monomer_prediction", "domain": "biology", "evaluated_system_id": "google-deepmind", "evaluated_system_name": "AlphaFold 3", "system_type": "structure_prediction_model", "model_or_version": "AlphaFold 3 FoldBench evaluation", "metric": "lddt", "score": 0.53, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "Chai-1", "baseline_score": 0.46, "evaluation_date": "2025-12", "benchmark_version": "FoldBench n=1522 assemblies", "dataset_split": "DNA monomer subset", "evaluator_type": "independent", "evaluator_name": "FoldBench authors", "reproducibility_status": "independently_reproduced", "claim_status": "reported", "source_tier": "A", "source_ids": [ "nature-foldbench-2025" ], "notes": "FoldBench reports AlphaFold 3 highest DNA monomer LDDT among five models." }, { "result_id": "foldbench-af3-rna-monomer-lddt-202512", "benchmark_id": "foldbench", "benchmark_name": "FoldBench", "task": "rna_monomer_prediction", "domain": "biology", "evaluated_system_id": "google-deepmind", "evaluated_system_name": "AlphaFold 3", "system_type": "structure_prediction_model", "model_or_version": "AlphaFold 3 FoldBench evaluation", "metric": "lddt", "score": 0.61, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "Protenix", "baseline_score": 0.59, "evaluation_date": "2025-12", "benchmark_version": "FoldBench n=1522 assemblies", "dataset_split": "RNA monomer subset", "evaluator_type": "independent", "evaluator_name": "FoldBench authors", "reproducibility_status": "independently_reproduced", "claim_status": "reported", "source_tier": "A", "source_ids": [ "nature-foldbench-2025" ], "notes": "FoldBench reports AlphaFold 3 highest RNA monomer LDDT among five models." }, { "result_id": "gnome-new-stable-crystals-202311", "benchmark_id": "gnome-materials-discovery", "benchmark_name": "GNoME materials discovery benchmark", "task": "stable_crystal_discovery", "domain": "materials", "evaluated_system_id": "google-deepmind", "evaluated_system_name": "GNoME", "system_type": "materials_discovery_model", "model_or_version": "GNoME Nature 2023", "metric": "new_stable_crystals_on_convex_hull", "score": 381000, "score_unit": "count", "higher_is_better": true, "baseline_name": "External stable crystal discoveries before GNoME", "baseline_score": 48000, "evaluation_date": "2023-11", "benchmark_version": "Nature 2023 dataset snapshot", "dataset_split": "updated convex hull", "evaluator_type": "peer_reviewed", "evaluator_name": "Nature peer-reviewed GNoME paper", "reproducibility_status": "partially_reproduced", "claim_status": "reported", "source_tier": "A", "source_ids": [ "nature-gnome-202311" ], "notes": "Paper reports 381,000 newly stable entries versus about 48,000 external stable crystals in the revised comparison." }, { "result_id": "alphaevolve-matrix-4x4-rank-202505", "benchmark_id": "alphaevolve-algorithm-discovery", "benchmark_name": "AlphaEvolve algorithm discovery tasks", "task": "4x4_complex_matrix_multiplication", "domain": "algorithmic science", "evaluated_system_id": "google-deepmind", "evaluated_system_name": "AlphaEvolve", "system_type": "coding_agent", "model_or_version": "AlphaEvolve May 2025", "metric": "scalar_multiplications", "score": 48, "score_unit": "multiplications", "higher_is_better": false, "baseline_name": "Strassen-derived 4x4 algorithm", "baseline_score": 49, "evaluation_date": "2025-05", "benchmark_version": "AlphaEvolve matrix multiplication tensor-rank task", "dataset_split": "4x4 complex-valued matrix multiplication", "evaluator_type": "self_reported", "evaluator_name": "Google DeepMind", "reproducibility_status": "partially_reproduced", "claim_status": "reported", "source_tier": "B", "source_ids": [ "official-alphaevolve-202505", "arxiv-alphaevolve-202506" ], "notes": "Objective is minimizing scalar multiplications; lower is better." }, { "result_id": "cfps-gpt5-ginkgo-cost-202602", "benchmark_id": "cfps-autonomous-lab-optimization", "benchmark_name": "CFPS autonomous-lab optimization", "task": "closed_loop_cell_free_protein_synthesis_optimization", "domain": "synthetic biology", "evaluated_system_id": "ginkgo-bioworks", "evaluated_system_name": "GPT-5 + Ginkgo Cloud Lab", "system_type": "closed_loop_autonomous_lab", "model_or_version": "GPT-5-driven Ginkgo Cloud Lab, February 2026", "metric": "sfgfp_production_cost", "score": 422, "score_unit": "usd_per_gram", "higher_is_better": false, "baseline_name": "State-of-the-art CFPS benchmark", "baseline_score": 698, "evaluation_date": "2026-02-05", "benchmark_version": "OpenAI/Ginkgo CFPS autonomous-lab optimization", "dataset_split": "36,000+ unique CFPS reaction compositions across 580 automated plates", "evaluator_type": "self_reported", "evaluator_name": "OpenAI / Ginkgo Bioworks", "reproducibility_status": "not_independently_reproduced", "claim_status": "reported", "source_tier": "B", "source_ids": [ "openai-ginkgo-cfps-autonomous-lab-202602", "ginkgo-openai-cfps-202602", "openai-ginkgo-cfps-preprint-202602", "biorxiv-openai-ginkgo-cfps-202602" ], "notes": "Closed-loop GPT-5 and Ginkgo Cloud Lab ran six optimization rounds with minimal human intervention; reported about 40% cost reduction and a 27% titer increase. Lower cost is better." }, { "result_id": "astabench-claude-opus-47-react-overall-202604", "benchmark_id": "astabench", "benchmark_name": "AstaBench", "task": "scientific_agent_research_suite", "domain": "scientific agents", "evaluated_system_id": "anthropic", "evaluated_system_name": "Claude Opus 4.7 with ReAct agent framework", "system_type": "frontier_model_agent", "model_or_version": "Claude Opus 4.7, extended thinking, April 2026 AstaBench update", "metric": "overall_accuracy", "score": 0.58, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "GPT-5.4 with ReAct agent framework", "baseline_score": 0.465, "evaluation_date": "2026-04-30", "benchmark_version": "AstaBench spring 2026 update", "dataset_split": "2,400+ scientific research problems across AstaBench categories", "evaluator_type": "independent", "evaluator_name": "Ai2", "reproducibility_status": "partially_reproduced", "claim_status": "reported", "source_tier": "B", "source_ids": [ "official-astabench", "ai2-astabench-update-202604" ], "notes": "Ai2 reports Claude Opus 4.7 ranking first overall at 58.0% with average cost of $3.54 per problem." }, { "result_id": "astabench-gpt-55-react-overall-202604", "benchmark_id": "astabench", "benchmark_name": "AstaBench", "task": "scientific_agent_research_suite", "domain": "scientific agents", "evaluated_system_id": "openai", "evaluated_system_name": "GPT-5.5 with ReAct agent framework", "system_type": "frontier_model_agent", "model_or_version": "GPT-5.5 xhigh reasoning, April 2026 AstaBench update", "metric": "overall_accuracy", "score": 0.529, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "GPT-5.4 with ReAct agent framework", "baseline_score": 0.465, "evaluation_date": "2026-04-30", "benchmark_version": "AstaBench spring 2026 update", "dataset_split": "2,400+ scientific research problems across AstaBench categories", "evaluator_type": "independent", "evaluator_name": "Ai2", "reproducibility_status": "partially_reproduced", "claim_status": "reported", "source_tier": "B", "source_ids": [ "official-astabench", "ai2-astabench-update-202604" ], "notes": "Ai2 reports GPT-5.5 at 52.9% and $1.61 per problem, behind Claude Opus 4.7 but on the quality-cost Pareto frontier." }, { "result_id": "arcagi2-gpt55-202605", "benchmark_id": "arc-agi-2", "benchmark_name": "ARC-AGI-2", "task": "abstraction_and_reasoning_novel_tasks", "domain": "fluid reasoning", "evaluated_system_id": "openai", "evaluated_system_name": "GPT-5.5", "system_type": "frontier_model", "model_or_version": "GPT-5.5, reported May 2026", "metric": "accuracy", "score": 0.85, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "Gemini 3.1 Pro", "baseline_score": 0.771, "evaluation_date": "2026-05", "benchmark_version": "ARC-AGI-2 public leaderboard", "dataset_split": "public evaluation set", "evaluator_type": "independent", "evaluator_name": "ARC Prize Foundation / LLM-Stats aggregation", "reproducibility_status": "partially_reproduced", "claim_status": "reported", "source_tier": "B", "source_ids": [ "arcprize-arc-agi2-leaderboard-202605", "llm-stats-arc-agi2-202605" ], "notes": "GPT-5.5 leads ARC-AGI-2 at 0.850; Gemini 3.1 Pro 0.771 and GPT-5.4 0.733 follow. Scores stay well below ARC-AGI-1 saturation, reflecting the benchmark's resistance to memorization." }, { "result_id": "arcagi2-gemini31pro-202605", "benchmark_id": "arc-agi-2", "benchmark_name": "ARC-AGI-2", "task": "abstraction_and_reasoning_novel_tasks", "domain": "fluid reasoning", "evaluated_system_id": "google-deepmind", "evaluated_system_name": "Gemini 3.1 Pro", "system_type": "frontier_model", "model_or_version": "Gemini 3.1 Pro, reported May 2026", "metric": "accuracy", "score": 0.771, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "GPT-5.4", "baseline_score": 0.733, "evaluation_date": "2026-05", "benchmark_version": "ARC-AGI-2 public leaderboard", "dataset_split": "public evaluation set", "evaluator_type": "independent", "evaluator_name": "ARC Prize Foundation / LLM-Stats aggregation", "reproducibility_status": "partially_reproduced", "claim_status": "reported", "source_tier": "B", "source_ids": [ "arcprize-arc-agi2-leaderboard-202605", "llm-stats-arc-agi2-202605" ], "notes": "Gemini 3.1 Pro scores 0.771 on ARC-AGI-2, second to GPT-5.5 and ahead of GPT-5.4 (0.733) and Claude Opus 4.6 (0.688)." }, { "result_id": "swebench-verified-claude-opus45-202603", "benchmark_id": "swe-bench-verified", "benchmark_name": "SWE-bench Verified", "task": "resolve_real_github_issues", "domain": "agentic software engineering", "evaluated_system_id": "anthropic", "evaluated_system_name": "Claude Opus 4.5", "system_type": "frontier_model", "model_or_version": "Claude Opus 4.5", "metric": "resolved_rate", "score": 0.809, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "Gemini 3.1 Pro", "baseline_score": 0.806, "evaluation_date": "2026-03", "benchmark_version": "SWE-bench Verified, 500 tasks", "dataset_split": "full verified split", "evaluator_type": "independent", "evaluator_name": "SWE-bench leaderboard aggregation", "reproducibility_status": "partially_reproduced", "claim_status": "reported", "source_tier": "C", "source_ids": [ "swebench-overview", "codeant-swebench-leaderboard-202602" ], "notes": "Claude Opus 4.5 (80.9%) leads a tight cluster with Opus 4.6 (80.8%) and Gemini 3.1 Pro (80.6%). Frontier models can reproduce some gold patches verbatim, indicating training-data contamination on this split." }, { "result_id": "swebench-pro-claude-opus45-202603", "benchmark_id": "swe-bench-pro", "benchmark_name": "SWE-bench Pro", "task": "resolve_real_multilanguage_github_issues", "domain": "agentic software engineering", "evaluated_system_id": "anthropic", "evaluated_system_name": "Claude Opus 4.5", "system_type": "frontier_model", "model_or_version": "Claude Opus 4.5, SEAL standardized scaffolding", "metric": "resolved_rate", "score": 0.459, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "Claude Opus 4.5 on SWE-bench Verified", "baseline_score": 0.809, "evaluation_date": "2026-03", "benchmark_version": "SWE-bench Pro, 1865 multi-language tasks", "dataset_split": "public SEAL leaderboard", "evaluator_type": "independent", "evaluator_name": "Scale AI SEAL", "reproducibility_status": "partially_reproduced", "claim_status": "reported", "source_tier": "C", "source_ids": [ "morphllm-swebench-pro-202603", "codeant-swebench-leaderboard-202602" ], "notes": "The same model drops from 80.9% on Verified to 45.9% on the contamination-resistant Pro split; the baseline shown is the model's own Verified score, to make the contamination gap explicit." }, { "result_id": "eci-gpt55-pro-202604", "benchmark_id": "epoch-ai-capabilities", "benchmark_name": "Epoch Capabilities Index", "task": "general_capability_index", "domain": "general model capability", "evaluated_system_id": "openai", "evaluated_system_name": "GPT-5.5 Pro", "system_type": "frontier_model", "model_or_version": "GPT-5.5 Pro, reported April 2026", "metric": "epoch_capabilities_index", "score": 159, "score_unit": "index_points", "higher_is_better": true, "baseline_name": "GPT-5 calibration anchor", "baseline_score": 150, "evaluation_date": "2026-04", "benchmark_version": "Epoch Capabilities Index", "dataset_split": "cross-benchmark composite", "evaluator_type": "independent", "evaluator_name": "Epoch AI", "reproducibility_status": "partially_reproduced", "claim_status": "reported", "source_tier": "B", "source_ids": [ "epoch-capabilities-index", "epoch-eci-gpt55pro-202604" ], "notes": "GPT-5.5 Pro set a new high of 159 (90% CI 156-162) on Epoch's cross-benchmark index, calibrated so GPT-5 = 150 and Claude 3.5 Sonnet = 130." }, { "result_id": "gpqa-diamond-gemini31pro-202605", "benchmark_id": "gpqa-diamond", "benchmark_name": "GPQA Diamond", "task": "graduate_level_science_qa", "domain": "scientific reasoning", "evaluated_system_id": "google-deepmind", "evaluated_system_name": "Gemini 3.1 Pro", "system_type": "frontier_model", "model_or_version": "Gemini 3.1 Pro, reported May 2026", "metric": "accuracy", "score": 0.943, "score_unit": "fraction", "higher_is_better": true, "baseline_name": "Expert human validator baseline", "baseline_score": 0.812, "evaluation_date": "2026-05", "benchmark_version": "GPQA Diamond", "dataset_split": "diamond subset", "evaluator_type": "independent", "evaluator_name": "Artificial Analysis / Stanford AI Index", "reproducibility_status": "partially_reproduced", "claim_status": "reported", "source_tier": "B", "source_ids": [ "artificial-analysis-intelligence-index", "stanford-ai-index-2026" ], "notes": "Frontier models now sit near 94% on GPQA Diamond, above the ~81.2% expert-validator baseline reported by the Stanford AI Index, indicating the knowledge benchmark is approaching saturation." } ], "clinical-trials": [ { "entity_id": "recursion-pharmaceuticals", "nct_id": "NCT05552755", "trial_id": "recursion-rec-4881-tupelo", "entity_name": "Recursion Pharmaceuticals", "linked_asset": "REC-4881", "phase": "Phase 1b/2", "status": "Recruiting", "indication": "Familial Adenomatous Polyposis", "sponsor": "Recursion Pharmaceuticals Inc.", "enrollment": 67, "primary_endpoint": "Part 1: Maximum (Peak) Plasma Drug Concentration (Cmax) of REC-4881", "start_date": "2023-07-10", "primary_completion_date": "2027-06", "completion_date": "2027-09", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-10", "source_ids": [ "clinicaltrials-nct05552755" ] }, { "entity_id": "iambic-therapeutics", "nct_id": "NCT06253871", "trial_id": "iambic-iam1363-her2-phase1", "entity_name": "Iambic Therapeutics", "linked_asset": "IAM1363", "phase": "Phase 1/1b", "status": "Recruiting", "indication": "Advanced HER2-altered cancers", "sponsor": "Iambic Therapeutics", "enrollment": 383, "primary_endpoint": "Number of participants with dose-limiting toxicities and adverse events", "start_date": "2024-03-25", "primary_completion_date": "2027-12", "completion_date": "2028-03", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-11", "source_ids": [ "clinicaltrials-nct06253871", "official-iambic-iam1363-first-patient-202403" ] }, { "entity_id": "insilico-medicine", "nct_id": "NCT05938920", "trial_id": "insilico-rentosertib-ipf-phase2a", "entity_name": "Insilico Medicine", "linked_asset": "ISM001-055 / rentosertib", "phase": "Phase 2a", "status": "Completed", "indication": "Idiopathic Pulmonary Fibrosis", "sponsor": "Insilico Medicine", "enrollment": 71, "primary_endpoint": "Change from baseline in forced vital capacity and safety/tolerability endpoints", "start_date": "2023-07-07", "primary_completion_date": "2024-11", "completion_date": "2025-03", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-11", "source_ids": [ "clinicaltrials-nct05938920", "nature-med-insilico-rentosertib-202506" ] }, { "entity_id": "generate-biomedicines", "nct_id": "NCT07359846", "trial_id": "generate-gb0895-solairia1", "entity_name": "Generate Biomedicines", "linked_asset": "GB-0895", "phase": "Phase 3", "status": "Recruiting", "indication": "Severe asthma", "sponsor": "Generate Biomedicines", "enrollment": 786, "primary_endpoint": "Annualized rate of clinically significant asthma exacerbations", "start_date": "2026-01-20", "primary_completion_date": "2028-03", "completion_date": "2028-03", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-11", "source_ids": [ "clinicaltrials-nct07359846", "official-generate-q1-202605" ] }, { "entity_id": "abcellera", "nct_id": "NCT04411628", "trial_id": "abcellera-lycov555-covid-phase1", "entity_name": "AbCellera", "linked_asset": "LY-CoV555 / bamlanivimab", "phase": "Phase 1", "status": "Completed", "indication": "COVID-19", "sponsor": "Eli Lilly and Company", "enrollment": 24, "primary_endpoint": "Safety and tolerability of LY-CoV555 in hospitalized participants with COVID-19", "start_date": "2020-06-08", "primary_completion_date": "2020-08", "completion_date": "2020-08", "source_tier": "A", "claim_status": "confirmed", "last_verified": "2026-05-11", "source_ids": [ "clinicaltrials-nct04411628", "official-abcellera-lilly-covid-202003" ] } ], "patents": [], "evidence-events": [ { "entity_id": "a-lab-lbnl", "event_id": "alab-nature-correction-2026-01", "entity_name": "A-Lab (LBNL)", "date": "2026-01-19", "evidence_type": "author_correction", "verification_status": "corrected_by_publisher", "replication_status": "post_publication_reanalysis", "negative_or_corrective_event": true, "retraction_or_correction": "correction", "claim_status": "corrected", "summary": "Nature Author Correction clarified that material novelty meant new to the prediction platform, not necessarily new to science, and reported 36 correct conclusions among 40 originally reported successes with 4 inconclusive compounds.", "last_verified": "2026-05-10", "source_tier": "A", "source_ids": [ "nature-alab-correction-202601" ] } ], "sources": [ { "id": "official-periodic-labs", "title": "Periodic Labs official site", "publisher": "Periodic Labs", "date": "2025", "url": "https://periodic.com", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-lila-sciences", "title": "Lila Sciences official site", "publisher": "Lila Sciences", "date": "2023", "url": "https://lila.ai", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-sakana-ai", "title": "Sakana AI official site", "publisher": "Sakana AI", "date": "2023", "url": "https://sakana.ai", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-futurehouse", "title": "FutureHouse official site", "publisher": "FutureHouse", "date": "2023", "url": "https://futurehouse.org", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-autoscience-institute", "title": "Autoscience Institute official site", "publisher": "Autoscience Institute", "date": "2023", "url": "https://autoscience.ai", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "academic_or_lab" }, { "id": "official-thesis", "title": "Thesis official site", "publisher": "Thesis", "date": "2025", "url": "https://www.ycombinator.com/companies/thesis", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-zeon-systems", "title": "Zeon Systems official site", "publisher": "Zeon Systems", "date": "2025", "url": "https://zeonsystems.ai", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-automata", "title": "Automata official site", "publisher": "Automata", "date": "2015", "url": "https://automata.tech", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-molecular-devices-automata-linq-202601", "title": "Molecular Devices Expands Access to Automated, AI-Ready Research Solutions Through Partnership with Automata's LINQ Platform", "publisher": "Molecular Devices", "date": "2026-01-29", "url": "https://www.moleculardevices.com/newsroom/news/molecular-devices-partnership-with-automata-linq-platform", "sourceType": "official", "accessed": "2026-05-13", "notes": "Molecular Devices announcement describing integration of its imaging and detection systems into Automata's LINQ platform for connected, AI-ready laboratory workflows.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-emerald-cloud-lab", "title": "Emerald Cloud Lab official site", "publisher": "Emerald Cloud Lab", "date": "2010", "url": "https://www.emeraldcloudlab.com", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-strateos", "title": "Strateos official site", "publisher": "Strateos", "date": "2012", "url": "https://strateos.com", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-chemify", "title": "Chemify official site", "publisher": "Chemify", "date": "2023", "url": "https://chemify.io", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-recursion-pharmaceuticals", "title": "Recursion Pharmaceuticals official site", "publisher": "Recursion Pharmaceuticals", "date": "2013", "url": "https://www.recursion.com", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-atinary", "title": "Atinary official site", "publisher": "Atinary", "date": "2019", "url": "https://atinary.com", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-atinary-boston-sdl-202602", "title": "Atinary Opens Self-Driving Labs in Boston, Bringing AI into the Physical World of R&D", "publisher": "Atinary", "date": "2026-03-04", "url": "https://atinary.com/news/opening-atinary-self-driving-labs-boston/", "sourceType": "official", "accessed": "2026-05-13", "notes": "Official Atinary announcement of the Boston self-driving lab, two Scientific Discovery Factories, closed-loop DMTA-L cycles, and robotics/instrument integration.", "source_tier": "B", "publisher_type": "company" }, { "id": "selectscience-atinary-boston-sdl-202602", "title": "Atinary launches Self-Driving Labs in Boston, bringing AI into the physical world of R&D", "publisher": "SelectScience", "date": "2026-02-11", "url": "https://www.selectscience.net/article/atinary-launches-self-driving-labs-r-in-boston-bringing-ai-into-the-physical-world-of-r-and-d", "sourceType": "press", "accessed": "2026-05-13", "notes": "Independent lab-automation coverage confirming Atinary's Boston lab, Scientific Discovery Factories, DMTA-L loop, and ABB/Agilent/Bruker/Chemspeed/Mettler-Toledo integration.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-ginkgo-bioworks", "title": "Ginkgo Bioworks official site", "publisher": "Ginkgo Bioworks", "date": "2008", "url": "https://www.ginkgobioworks.com", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "openai-ginkgo-cfps-autonomous-lab-202602", "title": "GPT-5 lowers the cost of cell-free protein synthesis", "publisher": "OpenAI", "date": "2026-02-05", "url": "https://openai.com/index/gpt-5-lowers-protein-synthesis-cost/", "sourceType": "official", "accessed": "2026-05-13", "notes": "OpenAI announcement describing a closed-loop GPT-5 and Ginkgo Cloud Lab system, 36,000+ CFPS reaction compositions across 580 automated plates, and lower sfGFP production cost.", "source_tier": "B", "publisher_type": "company" }, { "id": "ginkgo-openai-cfps-202602", "title": "Ginkgo Bioworks' Autonomous Laboratory Driven by OpenAI's GPT-5 Achieves 40% Improvement Over State-of-the-Art Scientific Benchmark", "publisher": "Ginkgo Bioworks", "date": "2026-02-05", "url": "https://www.prnewswire.com/news-releases/ginkgo-bioworks-autonomous-laboratory-driven-by-openais-gpt-5-achieves-40-improvement-over-state-of-the-art-scientific-benchmark-302680619.html", "sourceType": "official", "accessed": "2026-05-13", "notes": "Ginkgo announcement independently describing 36,000 experimental conditions across six closed-loop optimization cycles and Ginkgo's sale of the AI-improved reaction mix.", "source_tier": "B", "publisher_type": "company" }, { "id": "openai-ginkgo-cfps-preprint-202602", "title": "Using a GPT-5-driven autonomous lab to optimize the cost and titer of cell-free protein synthesis", "publisher": "OpenAI / Ginkgo Bioworks", "date": "2026-02-05", "url": "https://cdn.openai.com/pdf/5a12a3bc-96b7-4e07-9386-db6ee5bb2ed9/using-a-gpt-5-driven-autonomous-lab-to-optimize-the-cost-and-titer-of-cell-free-protein-synthesis.pdf", "sourceType": "research", "accessed": "2026-05-13", "notes": "OpenAI-hosted technical manuscript reporting the cost and titer results from the GPT-5-driven Ginkgo CFPS autonomous-lab study.", "source_tier": "B", "publisher_type": "preprint" }, { "id": "biorxiv-openai-ginkgo-cfps-202602", "title": "Using a GPT-5-driven autonomous lab to optimize the cost and titer of cell-free protein synthesis", "publisher": "bioRxiv", "date": "2026-02-05", "url": "https://www.biorxiv.org/content/10.64898/2026.02.05.703998v1", "sourceType": "research", "accessed": "2026-05-13", "notes": "bioRxiv preprint for the OpenAI/Ginkgo GPT-5 autonomous-lab CFPS study, reporting a 40% sfGFP production-cost reduction and 27% titer increase.", "source_tier": "B", "publisher_type": "preprint_server" }, { "id": "official-radical-ai", "title": "Radical AI official site", "publisher": "Radical AI", "date": "2022", "url": "https://radicalai.com", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-xaira-therapeutics", "title": "Xaira Therapeutics official site", "publisher": "Xaira Therapeutics", "date": "2024", "url": "https://www.xaira.com", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-a-lab-lbnl", "title": "A-Lab (LBNL) official site", "publisher": "A-Lab (LBNL)", "date": "2022", "url": "https://a-lab.lbl.gov", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "academic_or_lab" }, { "id": "official-acceleration-consortium", "title": "Acceleration Consortium official site", "publisher": "Acceleration Consortium", "date": "2021", "url": "https://acceleration.utoronto.ca", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-pnnl", "title": "PNNL official site", "publisher": "PNNL", "date": "1965", "url": "https://www.pnnl.gov", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "academic_or_lab" }, { "id": "official-google-deepmind", "title": "Google DeepMind official site", "publisher": "Google DeepMind", "date": "2010", "url": "https://deepmind.google", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-openai", "title": "OpenAI official site", "publisher": "OpenAI", "date": "2015", "url": "https://openai.com", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-isomorphic-labs", "title": "Isomorphic Labs official site", "publisher": "Isomorphic Labs", "date": "2021", "url": "https://www.isomorphiclabs.com", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-edison-scientific", "title": "Edison Scientific official site", "publisher": "Edison Scientific", "date": "2025", "url": "https://edisonscientific.com", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-insilico-medicine", "title": "Insilico Medicine official site", "publisher": "Insilico Medicine", "date": "2014", "url": "https://insilico.com", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-ami-labs", "title": "AMI Labs official site", "publisher": "AMI Labs", "date": "2025", "url": "https://amilabs.xyz", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-chemlex", "title": "ChemLex official site", "publisher": "ChemLex", "date": "2022", "url": "https://chemlex.com", "sourceType": "official", "accessed": "2026-04-24", "notes": "Primary organization website used for identity, product scope, and canonical URL.", "source_tier": "B", "publisher_type": "company" }, { "id": "artificial-analysis-leaderboard", "title": "LLM Leaderboard - Comparison of over 100 AI models", "publisher": "Artificial Analysis", "date": "2026-04", "url": "https://artificialanalysis.ai/leaderboards/models", "sourceType": "benchmark", "accessed": "2026-04-24", "notes": "Reference for model intelligence, price, speed, latency, and context-window UX patterns.", "source_tier": "B", "publisher_type": "benchmark_provider" }, { "id": "artificial-analysis-methodology", "title": "Artificial Analysis Benchmarking Methodology", "publisher": "Artificial Analysis", "date": "2026-03", "url": "https://artificialanalysis.ai/methodology", "sourceType": "methodology", "accessed": "2026-04-24", "notes": "Reference for live benchmarking, endpoint/provider definitions, and blended price methodology.", "source_tier": "B", "publisher_type": "methodology_provider" }, { "id": "artificial-analysis-intelligence-methodology", "title": "Artificial Analysis Intelligence Benchmarking Methodology", "publisher": "Artificial Analysis", "date": "2026-03", "url": "https://artificialanalysis.ai/methodology/intelligence-benchmarking", "sourceType": "methodology", "accessed": "2026-04-25", "notes": "Reference for the Intelligence Index evaluation mix across reasoning, coding, agentic work, knowledge, science, and instruction following.", "source_tier": "B", "publisher_type": "methodology_provider" }, { "id": "epoch-capabilities", "title": "AI Capabilities", "publisher": "Epoch AI", "date": "2026-04-18", "url": "https://epoch.ai/benchmarks", "sourceType": "benchmark", "accessed": "2026-04-24", "notes": "Reference for public AI capability tracking across 40+ benchmarks, including reasoning, agents, coding, science, and multimodal tasks.", "source_tier": "B", "publisher_type": "benchmark_provider" }, { "id": "epoch-capabilities-index", "title": "Epoch Capabilities Index", "publisher": "Epoch AI", "date": "2026-04", "url": "https://epoch.ai/benchmarks/eci", "sourceType": "benchmark", "accessed": "2026-04-25", "notes": "Reference for stitching many benchmark results into a general capability scale when individual benchmarks saturate.", "source_tier": "B", "publisher_type": "benchmark_provider" }, { "id": "epoch-benchmarking-about", "title": "About our benchmarking", "publisher": "Epoch AI", "date": "2026-04", "url": "https://epoch.ai/benchmarks/about", "sourceType": "methodology", "accessed": "2026-04-24", "notes": "Reference for internal and external benchmark source policy.", "source_tier": "B", "publisher_type": "methodology_provider" }, { "id": "metr-time-horizons", "title": "Measuring AI Ability to Complete Long Tasks", "publisher": "METR", "date": "2025-03-19", "url": "https://metr.org/blog/2025-03-19-measuring-ai-ability-to-complete-long-tasks/", "sourceType": "benchmark", "accessed": "2026-04-25", "notes": "Reference for task-completion time horizons and long-horizon autonomous-agent progress.", "source_tier": "B", "publisher_type": "benchmark_provider" }, { "id": "ai-2027-timelines", "title": "Timelines Forecast", "publisher": "AI 2027 / FutureSearch", "date": "2025-04", "url": "https://ai-2027.com/research/timelines-forecast", "sourceType": "forecast", "accessed": "2026-04-25", "notes": "Scenario reference for superhuman-coder and AI R&D acceleration trajectories; included as a scenario input, not an endorsement.", "source_tier": "C", "publisher_type": "forecast_provider" }, { "id": "openai-gpt-52", "title": "Introducing GPT-5.2", "publisher": "OpenAI", "date": "2025-12-11", "url": "https://openai.com/index/introducing-gpt-5-2", "sourceType": "benchmark", "accessed": "2026-04-25", "notes": "Reference for GPT-5.2 GDPval, SWE-bench Verified, GPQA Diamond, AIME 2025, ARC-AGI-1, and ARC-AGI-2 benchmark claims.", "source_tier": "B", "publisher_type": "benchmark_provider" }, { "id": "openai-gdpval", "title": "Measuring the performance of our models on real-world tasks", "publisher": "OpenAI", "date": "2025-09-25", "url": "https://openai.com/index/gdpval/", "sourceType": "benchmark", "accessed": "2026-04-25", "notes": "Reference for GDPval scope, occupational coverage, real-work deliverables, and one-shot workflow limitations.", "source_tier": "B", "publisher_type": "benchmark_provider" }, { "id": "metr-time-horizons-current", "title": "Task-Completion Time Horizons of Frontier AI Models", "publisher": "METR", "date": "2026-04-15", "url": "https://metr.org/time-horizons/", "sourceType": "benchmark", "accessed": "2026-04-25", "notes": "Reference for current time-horizon measurements, task distribution, and limits of well-specified software, ML, and cybersecurity tasks.", "source_tier": "B", "publisher_type": "benchmark_provider" }, { "id": "nature-hle", "title": "A benchmark of expert-level academic questions to assess AI capabilities", "publisher": "Nature", "date": "2026-01-28", "url": "https://www.nature.com/articles/s41586-025-09962-4", "sourceType": "benchmark", "accessed": "2026-04-25", "notes": "Reference for Humanity's Last Exam, benchmark saturation, multimodal expert questions, and calibration gaps.", "source_tier": "B", "publisher_type": "benchmark_provider" }, { "id": "swebench-overview", "title": "SWE-bench overview", "publisher": "SWE-bench", "date": "2026-04", "url": "https://www.swebench.com/SWE-bench/", "sourceType": "benchmark", "accessed": "2026-04-25", "notes": "Reference for SWE-bench as real-world GitHub issue patch generation with reproducible evaluation.", "source_tier": "B", "publisher_type": "benchmark_provider" }, { "id": "scivity-curation", "title": "Scivity Labs landscape curation", "publisher": "Scivity Labs", "date": "2026-04", "url": "https://scivity.org/landscape", "sourceType": "curation", "accessed": "2026-04-24", "notes": "Editorial synthesis and source registry for the AI-for-science landscape.", "source_tier": "B", "publisher_type": "curated_dataset" }, { "id": "official-helical", "title": "Helical official site", "publisher": "Helical", "date": "2026", "url": "https://www.helical.bio/", "sourceType": "official", "accessed": "2026-05-05", "notes": "Official product page for Helical's virtual AI lab, Model Factory, and pharma R&D positioning.", "source_tier": "B", "publisher_type": "company" }, { "id": "globenewswire-helical-seed-202604", "title": "Helical raises $10M for virtual AI lab that operates at pharma scale", "publisher": "GlobeNewswire / Helical", "date": "2026-04-14", "url": "https://www.globenewswire.com/news-release/2026/04/14/3273431/0/en/Helical-raises-10M-for-virtual-AI-lab-that-operates-at-pharma-scale-to-make-in-silico-discovery-reproducible.html", "sourceType": "press", "accessed": "2026-05-05", "notes": "Primary funding announcement with round size, investors, founders, founding date, and pharma deployment claims.", "source_tier": "B", "publisher_type": "company_press_release" }, { "id": "techeu-helical-seed-202604", "title": "Helical secures $10M to advance virtual AI lab for pharma research", "publisher": "Tech.eu", "date": "2026-04-14", "url": "https://tech.eu/2026/04/14/helical-secures-10m-to-advance-virtual-ai-lab-for-pharma-research/", "sourceType": "press", "accessed": "2026-05-05", "notes": "Independent confirmation of Helical's seed round, investor syndicate, founders, and pharma virtual-lab scope.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-fathom-therapeutics", "title": "Fathom Therapeutics official site", "publisher": "Fathom Therapeutics", "date": "2026", "url": "https://www.fathomtx.com/", "sourceType": "official", "accessed": "2026-05-05", "notes": "Official site used to verify Fathom Therapeutics' current brand and website.", "source_tier": "B", "publisher_type": "company" }, { "id": "prnewswire-fathom-series-a-202604", "title": "Fathom Therapeutics, Formerly Atommap, Raises $47 Million in Oversubscribed Series A Financing", "publisher": "PR Newswire / Atommap Corporation", "date": "2026-04-27", "url": "https://www.prnewswire.com/news-releases/fathom-therapeutics-formerly-atommap-raises-47-million-in-oversubscribed-series-a-financing-to-translate-physics-and-ai-enabled-small-molecule-design-into-next-generation-medicines-302754272.html", "sourceType": "press", "accessed": "2026-05-05", "notes": "Primary announcement for Fathom's Series A, founders, Microcosmos platform, headquarters, and discovery status.", "source_tier": "B", "publisher_type": "company_press_release" }, { "id": "cen-fathom-series-a-202604", "title": "Business Watch: Fathom Therapeutics raises $47 million for AI drug discovery", "publisher": "Chemical & Engineering News", "date": "2026-04", "url": "https://cen.acs.org/business/business-watch-braskem-new-majority-owner/104/web/2026/04", "sourceType": "industry-analysis", "accessed": "2026-05-05", "notes": "Chemistry-sector confirmation of Fathom's launch, Series A, founders, and Microcosmos drug-design engine.", "source_tier": "C", "publisher_type": "industry_analysis" }, { "id": "official-10x-science", "title": "10x Science official site", "publisher": "10x Science", "date": "2026", "url": "https://10xscience.com/", "sourceType": "official", "accessed": "2026-05-05", "notes": "Official company website used to verify the current brand and product positioning.", "source_tier": "B", "publisher_type": "company" }, { "id": "yc-10x-science-profile-202604", "title": "10x Science: The AI-native platform for next-generation protein characterization", "publisher": "Y Combinator", "date": "2026-04", "url": "https://www.ycombinator.com/companies/10x-science", "sourceType": "industry-analysis", "accessed": "2026-05-05", "notes": "Company profile used for founders, website, location, founding year, product scope, and launch details.", "source_tier": "C", "publisher_type": "industry_analysis" }, { "id": "techcrunch-10x-science-seed-202604", "title": "AI is spitting out more potential drugs than ever. This startup wants to figure out which ones matter.", "publisher": "TechCrunch", "date": "2026-04-22", "url": "https://techcrunch.com/2026/04/22/ai-is-spitting-out-more-potential-drugs-than-ever-this-start-up-wants-to-figure-out-which-ones-matter/", "sourceType": "press", "accessed": "2026-05-05", "notes": "Independent reporting on 10x Science's seed round, founders, founding date, and protein-characterization thesis.", "source_tier": "C", "publisher_type": "media" }, { "id": "prnewswire-10x-science-seed-202604", "title": "10x Science Raises $4.8M Seed to Build AI That Understands Proteins at the Molecular Level", "publisher": "PR Newswire / 10x Science", "date": "2026-04-22", "url": "https://www.prnewswire.com/news-releases/10x-science-raises-4-8m-seed-to-build-ai-that-understands-proteins-at-the-molecular-level-302750622.html", "sourceType": "press", "accessed": "2026-05-05", "notes": "Primary funding announcement for 10x Science's seed round, investor syndicate, and protein-characterization platform.", "source_tier": "B", "publisher_type": "company_press_release" }, { "id": "official-nvidia-lilly-lab-202601", "title": "NVIDIA and Lilly Announce Co-Innovation AI Lab to Reinvent Drug Discovery in the Age of AI", "publisher": "NVIDIA", "date": "2026-01-12", "url": "https://investor.nvidia.com/news/press-release-details/2026/NVIDIA-and-Lilly-Announce-Co-Innovation-AI-Lab-to-Reinvent-Drug-Discovery-in-the-Age-of-AI/default.aspx", "sourceType": "official", "accessed": "2026-05-05", "notes": "Primary source for the NVIDIA and Lilly AI co-innovation lab, investment scale, and drug-discovery scope.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-insilico-lilly-202603", "title": "Insilico Medicine Announces Global R&D Collaboration with Lilly", "publisher": "Insilico Medicine", "date": "2026-03-29", "url": "https://www.prnewswire.com/news-releases/insilico-medicine-announces-global-rd-collaboration-with-lilly-302727884.html", "sourceType": "press", "accessed": "2026-05-05", "notes": "Primary announcement for the Lilly and Insilico AI-driven R&D collaboration and deal terms.", "source_tier": "C", "publisher_type": "media" }, { "id": "techtarget-insilico-lilly-202603", "title": "Eli Lilly expands Insilico pact with $2.75B AI drug discovery deal", "publisher": "TechTarget Pharma Life Sciences", "date": "2026-03-30", "url": "https://www.techtarget.com/pharmalifesciences/news/366640790/Eli-Lilly-expands-Insilico-pact-with-275B-AI-drug-discovery-deal", "sourceType": "press", "accessed": "2026-05-05", "notes": "Independent pharma-industry coverage confirming the Lilly and Insilico deal value, upfront payment, and AI R&D scope.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-arpae-catalchem-e-projects-202604", "title": "CATALCHEM-E Project Descriptions", "publisher": "ARPA-E", "date": "2026-04", "url": "https://arpa-e.energy.gov/sites/default/files/2026-04/CATALCHEM-E_selection_project%20descriptions_final_1.pdf", "sourceType": "official", "accessed": "2026-05-05", "notes": "Official project-description PDF for the $34M CATALCHEM-E autonomous-lab catalyst-development awards.", "source_tier": "A", "publisher_type": "government" }, { "id": "ncsu-catalchem-e-award-202604", "title": "NC State Receives $3M from ARPA-E to Advance Self-Driving Laboratory Research", "publisher": "NC State University College of Engineering", "date": "2026-04-20", "url": "https://engr.ncsu.edu/news/2026/04/20/nc-state-receives-3m-from-arpa-e-to-advance-self-driving-laboratory-research/", "sourceType": "official", "accessed": "2026-05-05", "notes": "University confirmation of CATALCHEM-E award size, self-driving-lab scope, and 12-project program context.", "source_tier": "B", "publisher_type": "academic_or_lab" }, { "id": "ames-catalchem-e-award-202604", "title": "Ames National Laboratory selected for ARPA-E funding to advance AI-driven catalyst development", "publisher": "Ames National Laboratory", "date": "2026-04-13", "url": "https://www.ameslab.gov/news/ames-national-laboratory-selected-for-arpa-e-funding-to-advance-ai-driven-catalyst-development", "sourceType": "official", "accessed": "2026-05-13", "notes": "Ames confirmation of its CATALCHEM-E A-TEAM award, robotic synthesis/testing workflow, and collaborators Worcester Polytechnic Institute, Chemspeed, and BASF.", "source_tier": "B", "publisher_type": "academic_or_lab" }, { "id": "p2science-catalchem-e-award-202604", "title": "P2 Science Selected to Receive $2.8m in Federal Funding for AI-Enhanced Autonomous Labs", "publisher": "P2 Science", "date": "2026-04-13", "url": "https://www.prnewswire.com/news-releases/p2-science-selected-to-receive-2-8m-in-federal-funding-for-new-technologies-that-aim-to-accelerate-catalyst-innovation-through-ai-enhanced-autonomous-labs-302739881.html", "sourceType": "official", "accessed": "2026-05-13", "notes": "Company announcement confirming P2 Science's CATALCHEM-E HEAT FACTORY award, autonomous robotics and machine-learning catalyst-discovery scope, and Matter Lab / National Labs of the Rockies collaborators.", "source_tier": "B", "publisher_type": "company" }, { "id": "techcrunch-anthropic-coefficient-bio-202604", "title": "Anthropic buys biotech startup Coefficient Bio in $400M deal: Reports", "publisher": "TechCrunch", "date": "2026-04-03", "url": "https://techcrunch.com/2026/04/03/anthropic-buys-biotech-startup-coefficient-bio-in-400m-deal-reports/", "sourceType": "press", "accessed": "2026-05-05", "notes": "Independent confirmation that Anthropic closed the Coefficient Bio acquisition and folded the team into life sciences.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-proqr-ginkgo-nebula-202604", "title": "ProQR Announces Partnership with Ginkgo Bioworks and Formation of AI Advisory Board", "publisher": "ProQR Therapeutics", "date": "2026-04-08", "url": "https://www.proqr.com/press-releases/proqr-announces-partnership-with-ginkgo-bioworks-and-formation-of-ai-advisory-board", "sourceType": "official", "accessed": "2026-05-05", "notes": "Primary partnership announcement covering Ginkgo Nebula access, ProQR's AI advisory board, and clinical-timeline claims.", "source_tier": "B", "publisher_type": "company" }, { "id": "rdworld-proqr-ginkgo-nebula-202604", "title": "ProQR turns to Ginkgo's autonomous lab to scale AI-enabled RNA editing discovery", "publisher": "R&D World", "date": "2026-04-14", "url": "https://www.rdworldonline.com/proqr-turns-to-ginkgos-autonomous-lab-to-scale-ai-enabled-rna-editing-discovery/", "sourceType": "press", "accessed": "2026-05-05", "notes": "Industry coverage confirming the ProQR and Ginkgo autonomous-lab partnership and development-candidate timeline.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-aria-activation-partners-202604", "title": "Become an Activation Partner", "publisher": "ARIA", "date": "2026-04", "url": "https://aria.org.uk/activation-partners/become-an-activation-partner/", "sourceType": "official", "accessed": "2026-05-05", "notes": "Official ARIA call for 100M GBP of Activation Partners, including AI-for-science tools, AI Scientist systems, and autonomous labs.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-doe-genesis-mission-rfa-202603", "title": "Energy Department Announces $293 Million in Funding to Support Genesis Mission National Science and Technology Challenges", "publisher": "U.S. Department of Energy", "date": "2026-03-17", "url": "https://www.energy.gov/articles/energy-department-announces-293-million-funding-support-genesis-mission-national-science", "sourceType": "official", "accessed": "2026-05-05", "notes": "Primary DOE source for the Genesis Mission RFA, funding amount, national challenge areas, and AI-for-science program structure.", "source_tier": "A", "publisher_type": "government" }, { "id": "official-futurehouse-disco-202604", "title": "DISCO: Inventing Enzymes for Chemistry that Nature Never Explored", "publisher": "FutureHouse", "date": "2026-04-08", "url": "https://www.futurehouse.org/research-announcements/disco-enzymes", "sourceType": "official", "accessed": "2026-05-05", "notes": "FutureHouse announcement for the DISCO enzyme-design model, collaborators, benchmark claims, and wet-lab validation results.", "source_tier": "B", "publisher_type": "company" }, { "id": "arxiv-disco-enzyme-design-202604", "title": "General Multimodal Protein Design Enables DNA-Encoding of Chemistry", "publisher": "arXiv", "date": "2026-04-06", "url": "https://arxiv.org/abs/2604.05181", "sourceType": "research", "accessed": "2026-05-05", "notes": "Research preprint for DISCO, including sequence-structure co-design methods and new-to-nature enzyme validation.", "source_tier": "B", "publisher_type": "preprint_server" }, { "id": "official-sakana-fugu-202604", "title": "Sakana Fugu: A Multi-Agent Orchestration System as a Foundation Model", "publisher": "Sakana AI", "date": "2026-04-24", "url": "https://sakana.ai/fugu-beta/", "sourceType": "official", "accessed": "2026-05-05", "notes": "Sakana AI announcement for the Fugu beta, multi-agent orchestration system, and scientific-reasoning positioning.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-futurehouse-bixbench-202503", "title": "Announcing BixBench: A Benchmark to Evaluate AI Agents on Bioinformatics Tasks", "publisher": "FutureHouse", "date": "2025-03-04", "url": "https://www.futurehouse.org/research-announcements/bixbench", "sourceType": "official", "accessed": "2026-05-05", "notes": "Primary benchmark announcement for BixBench's real-world bioinformatics scenarios and autonomous-agent evaluation design.", "source_tier": "B", "publisher_type": "company" }, { "id": "arxiv-bixbench-202503", "title": "BixBench: a Comprehensive Benchmark for LLM-based Agents in Computational Biology", "publisher": "arXiv", "date": "2025-02", "url": "https://arxiv.org/abs/2503.00096", "sourceType": "research", "accessed": "2026-05-05", "notes": "Research preprint describing BixBench's task set, agent framework, and baseline model results.", "source_tier": "B", "publisher_type": "preprint_server" }, { "id": "official-edison-labbench2-202605", "title": "LABBench2: An Improved Benchmark for Measuring AI in Biology Research", "publisher": "Edison Scientific", "date": "2026-05-02", "url": "https://edisonscientific.com/articles/labbench2-an-improved-benchmark", "sourceType": "official", "accessed": "2026-05-05", "notes": "Edison Scientific announcement for LABBench2's launch, 1,892-task scope, and practical biology-research evaluation design.", "source_tier": "B", "publisher_type": "company" }, { "id": "huggingface-labbench2-202603", "title": "EdisonScientific/labbench2", "publisher": "Hugging Face", "date": "2026-03", "url": "https://huggingface.co/datasets/EdisonScientific/labbench2", "sourceType": "research", "accessed": "2026-05-05", "notes": "Dataset card for LABBench2, including task count, benchmark description, harness availability, and March data correction.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "id": "official-anthropic-biomysterybench-202604", "title": "Evaluating Claude's bioinformatics research capabilities with BioMysteryBench", "publisher": "Anthropic", "date": "2026-04-29", "url": "https://www.anthropic.com/research/Evaluating-Claude-For-Bioinformatics-With-BioMysteryBench", "sourceType": "official", "accessed": "2026-05-05", "notes": "Anthropic research post introducing BioMysteryBench, its 99 bioinformatics tasks, expert baselines, and reliability analysis.", "source_tier": "B", "publisher_type": "company" }, { "id": "huggingface-biomysterybench-202604", "title": "Anthropic/BioMysteryBench-full", "publisher": "Hugging Face", "date": "2026-04", "url": "https://huggingface.co/datasets/Anthropic/BioMysteryBench-full", "sourceType": "research", "accessed": "2026-05-05", "notes": "Dataset card for the full BioMysteryBench problem set, data files, and human-solvable labels.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "id": "official-recursion-q1-2026", "title": "Recursion Reports First Quarter Financial Results and Provides Business Update", "publisher": "Recursion Pharmaceuticals", "date": "2026-05-06", "url": "https://www.globenewswire.com/news-release/2026/05/06/3288617/0/en/Recursion-Reports-First-Quarter-Financial-Results-and-Provides-Business-Update.html", "sourceType": "official", "accessed": "2026-05-08", "notes": "Recursion Q1 2026 results: REC-4881 FAP proof of concept, fifth Sanofi milestone, REC-1245 DAHLIA Phase 1 data, runway to early 2028.", "source_tier": "B", "publisher_type": "company" }, { "id": "fool-recursion-q1-2026", "title": "Recursion (RXRX) Q1 2026 Earnings Transcript", "publisher": "The Motley Fool", "date": "2026-05-07", "url": "https://www.fool.com/earnings/call-transcripts/2026/05/07/recursion-rxrx-q1-2026-earnings-transcript/", "sourceType": "press", "accessed": "2026-05-08", "notes": "Independent transcript of Recursion's Q1 2026 earnings call covering REC-4881 FAP results, Sanofi milestone, and pipeline progress.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-ginkgo-q1-2026", "title": "Ginkgo Bioworks Reports First Quarter 2026 Financial Results, Completes Divestiture of Biosecurity and Continues to Scale Autonomous Lab", "publisher": "Ginkgo Bioworks", "date": "2026-05-07", "url": "https://www.prnewswire.com/news-releases/ginkgo-bioworks-reports-first-quarter-2026-financial-results-completes-divestiture-of-biosecurity-and-continues-to-scale-autonomous-lab-302766171.html", "sourceType": "official", "accessed": "2026-05-13", "notes": "Ginkgo Q1 2026 results: $19.5M revenue, $143.9M cash and cash equivalents, $229.6M marketable securities, Biosecurity divestiture closed April 3, Cloud Lab traction with ProQR and Amazon Bio Discovery, plans to double Nebula in 2026.", "source_tier": "B", "publisher_type": "company" }, { "id": "stocktitan-ginkgo-q1-2026", "title": "Ginkgo Bioworks Q1 2026 results call set for May 7", "publisher": "StockTitan", "date": "2026-04", "url": "https://www.stocktitan.net/news/DNA/ginkgo-bioworks-announces-date-of-first-quarter-2026-results-6f4hcokah00d.html", "sourceType": "press", "accessed": "2026-05-08", "notes": "Independent confirmation of Ginkgo's Q1 2026 results call date and Biosecurity divestiture timing.", "source_tier": "C", "publisher_type": "media" }, { "id": "openai-genebench-202604", "title": "GeneBench: A benchmark for AI agents on multi-stage scientific data analysis in genetics and quantitative biology", "publisher": "OpenAI", "date": "2026-04-23", "url": "https://cdn.openai.com/pdf/6dc7175d-d9e7-4b8d-96b8-48fe5798cd5b/oai_genebench_benchmark.pdf", "sourceType": "benchmark", "accessed": "2026-05-08", "notes": "OpenAI's primary benchmark report for GeneBench, including 103 multi-stage genomics and quantitative biology tasks across 10 domains.", "source_tier": "B", "publisher_type": "benchmark_provider" }, { "id": "biorxiv-genebench-202604", "title": "GeneBench: Assessing AI Agents for Multi-Stage Inference Problems in Genomics and Quantitative Biology", "publisher": "bioRxiv", "date": "2026-04-22", "url": "https://www.biorxiv.org/content/10.64898/2026.04.22.720113v1", "sourceType": "research", "accessed": "2026-05-08", "notes": "Independent preprint accompanying the GeneBench release covering benchmark design, scoring, and current model results.", "source_tier": "B", "publisher_type": "preprint_server" }, { "id": "openai-gpt-rosalind-202604", "title": "Introducing GPT-Rosalind for life sciences research", "publisher": "OpenAI", "date": "2026-04-16", "url": "https://openai.com/index/introducing-gpt-rosalind/", "sourceType": "official", "accessed": "2026-05-08", "notes": "OpenAI's announcement of GPT-Rosalind, a frontier reasoning model for biology, drug discovery, and translational medicine, plus a Codex life-sciences plugin.", "source_tier": "B", "publisher_type": "company" }, { "id": "fiercebiotech-gpt-rosalind-202604", "title": "OpenAI launches biotech-specific AI model dubbed GPT-Rosalind", "publisher": "Fierce Biotech", "date": "2026-04-16", "url": "https://www.fiercebiotech.com/biotech/openai-launches-biotech-specific-ai-model-gpt-rosalind", "sourceType": "press", "accessed": "2026-05-08", "notes": "Independent biotech press confirmation of GPT-Rosalind's launch, target customer set, and trusted access model.", "source_tier": "C", "publisher_type": "media" }, { "id": "the-decoder-gpt-rosalind-bixbench-202604", "title": "OpenAI launches GPT-Rosalind, a reasoning model built for life sciences research", "publisher": "The Decoder", "date": "2026-04-17", "url": "https://the-decoder.com/openai-launches-gpt-rosalind-a-reasoning-model-built-for-life-sciences-research/", "sourceType": "press", "accessed": "2026-05-13", "notes": "Secondary coverage transcribing OpenAI's BixBench Pass@1 chart numbers for GPT-Rosalind, GPT-5.4, and other comparison models.", "source_tier": "C", "publisher_type": "media" }, { "id": "creati-isomorphic-trials-202604", "title": "Isomorphic Labs to Begin Human Trials of AI-Designed Drugs From DeepMind Spinoff", "publisher": "Creati", "date": "2026-04-25", "url": "https://creati.ai/ai-news/2026-04-25/isomorphic-labs-ai-designed-drugs-human-trials/", "sourceType": "press", "accessed": "2026-05-08", "notes": "Coverage of Isomorphic Labs' April 16 announcement at WIRED Health that the company is gearing up to enter the clinic with its first AI-designed cancer drugs.", "source_tier": "C", "publisher_type": "media" }, { "id": "clinicaltrialsarena-isomorphic-trials-202604", "title": "Isomorphic Labs prepares to launch trials for AI-designed drugs", "publisher": "Clinical Trials Arena", "date": "2026-04", "url": "https://www.clinicaltrialsarena.com/news/isomorphic-labs-prepares-trials-ai-designed-drugs/", "sourceType": "press", "accessed": "2026-05-08", "notes": "Independent clinical-trial industry coverage of Isomorphic's preparation for first-in-human studies of its AlphaFold-derived oncology candidates.", "source_tier": "C", "publisher_type": "media" }, { "id": "blog-google-deep-research-max-202604", "title": "Deep Research Max: a step change for autonomous research agents", "publisher": "Google", "date": "2026-04-21", "url": "https://blog.google/innovation-and-ai/models-and-research/gemini-models/next-generation-gemini-deep-research/", "sourceType": "official", "accessed": "2026-05-08", "notes": "Google's announcement of Deep Research and Deep Research Max, autonomous research agents built on Gemini 3.1 Pro with MCP tool support and ~160 queries per task.", "source_tier": "B", "publisher_type": "company" }, { "id": "edtechinnovationhub-deep-research-max-202604", "title": "Google DeepMind launches Deep Research Max autonomous AI research agent", "publisher": "EdTech Innovation Hub", "date": "2026-04-21", "url": "https://www.edtechinnovationhub.com/news/googles-new-ai-agent-will-run-160-searches-while-you-sleep-and-hand-you-the-report-by-morning", "sourceType": "press", "accessed": "2026-05-08", "notes": "Independent coverage of Deep Research and Deep Research Max launch on Gemini 3.1 Pro and the autonomous-research workflow positioning.", "source_tier": "C", "publisher_type": "media" }, { "id": "axios-gpt-rosalind-202604", "title": "OpenAI launches new AI model for life sciences research", "publisher": "Axios", "date": "2026-04-16", "url": "https://www.axios.com/2026/04/16/openai-models-life-sciences-drugs", "sourceType": "press", "accessed": "2026-05-08", "notes": "Independent confirmation of GPT-Rosalind's launch, partner set, and trusted-access deployment.", "source_tier": "C", "publisher_type": "media" }, { "id": "cmu-coscientist-202312", "title": "CMU-Designed Artificially Intelligent Coscientist Automates Scientific Discovery", "publisher": "Carnegie Mellon University", "date": "2023-12-20", "url": "https://www.cmu.edu/news/stories/archives/2023/december/cmu-designed-artificially-intelligent-coscientist-automates-scientific-discovery", "sourceType": "official", "accessed": "2026-05-09", "notes": "University announcement for Coscientist, Emerald Cloud Lab execution, and autonomous chemistry scope.", "source_tier": "B", "publisher_type": "academic_or_lab" }, { "id": "nature-coscientist-202312", "title": "Autonomous chemical research with large language models", "publisher": "Nature", "date": "2023-12-20", "url": "https://www.nature.com/articles/s41586-023-06792-0", "sourceType": "research", "accessed": "2026-05-09", "notes": "Peer-reviewed Coscientist paper describing autonomous design, planning, and execution of chemistry experiments.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "id": "official-kebotix", "title": "Kebotix official site", "publisher": "Kebotix", "date": "2026", "url": "https://www.kebotix.com/", "sourceType": "official", "accessed": "2026-05-09", "notes": "Primary site for Kebotix identity and self-driving materials-discovery positioning.", "source_tier": "B", "publisher_type": "company" }, { "id": "finsmes-kebotix-series-a-202004", "title": "Kebotix Raises $11.4M in Series A Funding", "publisher": "FinSMEs", "date": "2020-04-16", "url": "https://www.finsmes.com/2020/04/kebotix-raises-11-4m-in-series-a-funding.html", "sourceType": "press", "accessed": "2026-05-09", "notes": "Independent funding coverage for Kebotix Series A and self-driving lab scope.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-citrine-informatics", "title": "Citrine Informatics official site", "publisher": "Citrine Informatics", "date": "2026", "url": "https://citrine.io/", "sourceType": "official", "accessed": "2026-05-09", "notes": "Primary source for Citrine identity and materials AI platform positioning.", "source_tier": "B", "publisher_type": "company" }, { "id": "businesswire-citrine-series-c-202301", "title": "Citrine Informatics Raises $16M in Series C Financing", "publisher": "BusinessWire / Citrine Informatics", "date": "2023-01-04", "url": "https://www.businesswire.com/news/home/20230104005839/en/Citrine-Informatics-Raises-%2416M-in-Series-C-Financing", "sourceType": "press", "accessed": "2026-05-09", "notes": "Funding and commercial-context source for Citrine materials and chemicals AI platform.", "source_tier": "B", "publisher_type": "company_press_release" }, { "id": "official-artificial", "title": "Artificial official site", "publisher": "Artificial", "date": "2026", "url": "https://www.artificial.com/", "sourceType": "official", "accessed": "2026-05-09", "notes": "Primary product source for Artificial lab orchestration, digital twins, and AI agents.", "source_tier": "B", "publisher_type": "company" }, { "id": "techcrunch-artificial-series-a-202105", "title": "Artificial raises $21M led by Microsoft M12 for a lab automation platform", "publisher": "TechCrunch", "date": "2021-05-18", "url": "https://techcrunch.com/2021/05/18/artificial-raises-21m-led-by-microsofts-m12-for-a-lab-automation-platform-aimed-at-life-sciences-rd/", "sourceType": "press", "accessed": "2026-05-09", "notes": "Independent reporting on Artificial funding, aLab Suite, customers, and lab-automation positioning.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-synthace", "title": "Synthace platform", "publisher": "Synthace", "date": "2026", "url": "https://www.synthace.com/platform", "sourceType": "official", "accessed": "2026-05-09", "notes": "Primary source for Synthace experiment design, automation instruction, and structured data workflow.", "source_tier": "B", "publisher_type": "company" }, { "id": "finsmes-synthace-series-c-202111", "title": "Synthace Raises $35M in Series C Funding", "publisher": "FinSMEs", "date": "2021-11-16", "url": "https://www.finsmes.com/2021/11/synthace-raises-35m-in-series-c-funding.html", "sourceType": "press", "accessed": "2026-05-09", "notes": "Independent funding source for Synthace Series C and R&D cloud positioning.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-tetrascience", "title": "Tetra OS official site", "publisher": "TetraScience", "date": "2026", "url": "https://www.tetrascience.com/", "sourceType": "official", "accessed": "2026-05-09", "notes": "Primary source for TetraScience scientific data and AI platform positioning.", "source_tier": "B", "publisher_type": "company" }, { "id": "fierce-tetrascience-series-b-202104", "title": "TetraScience scores $80M series B investment, rolls out open R&D data cloud", "publisher": "Fierce Healthcare", "date": "2021-04-16", "url": "https://www.fiercehealthcare.com/tech/tetrascience-scores-80m-series-b-investment-rolls-out-open-r-d-data-cloud", "sourceType": "press", "accessed": "2026-05-09", "notes": "Independent coverage of TetraScience funding, R&D Data Cloud, and enterprise pharma use.", "source_tier": "C", "publisher_type": "media" }, { "id": "siemens-riffyn-x", "title": "Riffyn X enterprise recipe management", "publisher": "Siemens Software", "date": "2026", "url": "https://plm.sw.siemens.com/en-US/riffyn-x/", "sourceType": "official", "accessed": "2026-05-09", "notes": "Current product source for Riffyn X process design and data platform.", "source_tier": "B", "publisher_type": "company" }, { "id": "biospace-riffyn-series-b-201905", "title": "Riffyn Raises a $15M Series B to Accelerate Its Growth in Life Sciences R&D", "publisher": "BioSpace", "date": "2019-05-29", "url": "https://www.biospace.com/riffyn-raises-a-15m-series-b-to-accelerate-its-growth-in-life-sciences-r-and-amp-d", "sourceType": "press", "accessed": "2026-05-09", "notes": "Independent funding source for Riffyn platform scope, investors, and R&D data analytics claims.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-opentrons-flex", "title": "Opentrons Flex", "publisher": "Opentrons", "date": "2026", "url": "https://opentrons.com/robots/flex", "sourceType": "official", "accessed": "2026-05-09", "notes": "Primary source for Opentrons Flex, open-source API, modular robotics, and AI-agent compatibility.", "source_tier": "B", "publisher_type": "company" }, { "id": "techcrunch-opentrons-flex-202305", "title": "Opentrons aims to democratize lab access with its Flex robot", "publisher": "TechCrunch", "date": "2023-05-22", "url": "https://techcrunch.com/2023/05/22/opentrons-aims-to-democratize-lab-access-with-its-flex-robot/", "sourceType": "press", "accessed": "2026-05-09", "notes": "Independent coverage of Opentrons Flex and reproducible programmable experimentation thesis.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-tecan-history", "title": "Tecan company history", "publisher": "Tecan", "date": "2026", "url": "https://www.tecan.com/history", "sourceType": "official", "accessed": "2026-05-09", "notes": "Primary source for Tecan founding, history, and lab-automation product lineage.", "source_tier": "B", "publisher_type": "company" }, { "id": "stockanalysis-tecan-profile-202603", "title": "Tecan Group AG Company Profile and Description", "publisher": "StockAnalysis", "date": "2026-03", "url": "https://stockanalysis.com/quote/vie/TECN/company/", "sourceType": "industry-analysis", "accessed": "2026-05-09", "notes": "Independent company profile for Tecan headquarters, founding year, public listing, and product scope.", "source_tier": "C", "publisher_type": "industry_analysis" }, { "id": "official-highres", "title": "HighRes official site", "publisher": "HighRes Biosolutions", "date": "2026", "url": "https://www.highres.com/", "sourceType": "official", "accessed": "2026-05-09", "notes": "Primary source for HighRes lab automation, Cellario, Nucleus, and application areas.", "source_tier": "B", "publisher_type": "company" }, { "id": "biodev-highres-cellario-202509", "title": "Cellario (HighRes Biosolutions)", "publisher": "Bio.Dev", "date": "2025-09-08", "url": "https://bio.dev/tools/cellario-highres-biosolutions", "sourceType": "industry-analysis", "accessed": "2026-05-09", "notes": "Independent tool profile for Cellario orchestration, scheduling, and API-first lab automation.", "source_tier": "C", "publisher_type": "industry_analysis" }, { "id": "official-generate-biomedicines", "title": "Generate Biomedicines official site", "publisher": "Generate Biomedicines", "date": "2026", "url": "https://generatebiomedicines.com/", "sourceType": "official", "accessed": "2026-05-09", "notes": "Primary source for Generate Biomedicines platform, pipeline, and generative biology positioning.", "source_tier": "B", "publisher_type": "company" }, { "id": "fierce-generate-series-b-202111", "title": "Flagship-founded Generate snags $370M to generate medicines based on proteins", "publisher": "Fierce Biotech", "date": "2021-11-18", "url": "https://www.fiercebiotech.com/biotech/flagship-founded-generate-snags-370m-to-generate-medicines-based-proteins-building-blocks", "sourceType": "press", "accessed": "2026-05-09", "notes": "Independent coverage of Generate Series B and protein-generation platform.", "source_tier": "C", "publisher_type": "media" }, { "id": "businesswire-generate-series-c-202309", "title": "Generate Biomedicines Announces Close of $273M Series C Financing", "publisher": "BusinessWire / Generate Biomedicines", "date": "2023-09-14", "url": "https://www.businesswire.com/news/home/20230914742403/en/GenerateBiomedicines-Announces-Close-of-%24273M-Series-C-Financing-to-Advance-Its-Generative-AI-Pipeline-of-Preclinical-and-Clinical-Protein-Therapeutics", "sourceType": "press", "accessed": "2026-05-09", "notes": "Funding and pipeline source for Generate Series C, program count, and investor syndicate.", "source_tier": "B", "publisher_type": "company_press_release" }, { "id": "official-insitro-purpose", "title": "Better Medicine in the Epoch of Machine Learning", "publisher": "Insitro", "date": "2026", "url": "https://www.insitro.com/purpose/", "sourceType": "official", "accessed": "2026-05-09", "notes": "Primary source for Insitro founding year, mission, headquarters, and ML drug discovery platform.", "source_tier": "B", "publisher_type": "company" }, { "id": "ap-insitro-ai-drug-discovery-202412", "title": "Better drugs through AI? Insitro CEO on what machine learning can teach Big Pharma", "publisher": "Associated Press", "date": "2024-12-02", "url": "https://apnews.com/article/004c0ce0442b72c37bfec6e032796808", "sourceType": "press", "accessed": "2026-05-09", "notes": "Independent profile of Insitro, Daphne Koller, ML drug discovery, and pharma partnerships.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-cellarity-about", "title": "Cellarity about", "publisher": "Cellarity", "date": "2026", "url": "https://cellarity.com/about/", "sourceType": "official", "accessed": "2026-05-09", "notes": "Primary source for Cellarity platform, founders, and cell-state-correcting medicine thesis.", "source_tier": "B", "publisher_type": "company" }, { "id": "cellarity-series-c-202210", "title": "Cellarity Announces Close of $121 Million Series C Financing", "publisher": "Cellarity", "date": "2022-10-04", "url": "https://cellarity.com/news_item/cellarity-announces-close-of-121-million-series-c-financing/", "sourceType": "press", "accessed": "2026-05-09", "notes": "Funding source for Cellarity Series C, total funding, investors, and platform context.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-abcellera-platform", "title": "AbCellera platform", "publisher": "AbCellera", "date": "2026", "url": "https://www.abcellera.com/platform", "sourceType": "official", "accessed": "2026-05-09", "notes": "Primary source for AbCellera target-to-clinic antibody discovery and manufacturing platform.", "source_tier": "B", "publisher_type": "company" }, { "id": "renaissance-abcellera-ipo-202011", "title": "AI drug discovery platform AbCellera files for a $200 million IPO", "publisher": "Renaissance Capital", "date": "2020-11-20", "url": "https://www.renaissancecapital.com/IPO-Center/News/73406/AI-drug-discovery-platform-AbCellera-files-for-a-%24200-million-IPO", "sourceType": "press", "accessed": "2026-05-09", "notes": "Independent source for AbCellera founding year, AI-powered antibody platform, and IPO context.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-iambic-platform", "title": "Iambic Therapeutics platform", "publisher": "Iambic Therapeutics", "date": "2026", "url": "https://www.iambic.ai/platform", "sourceType": "official", "accessed": "2026-05-09", "notes": "Primary source for Iambic Enchant, NeuralPLexer, automation, and weekly data-loop claims.", "source_tier": "B", "publisher_type": "company" }, { "id": "fierce-iambic-series-b-202310", "title": "Nearing clinic, AI-driven Iambic posts $100M round, Nvidia pact", "publisher": "Fierce Biotech", "date": "2023-10-04", "url": "https://www.fiercebiotech.com/biotech/iambic-ups-rhythm-ai-enabled-race-clinic-revealing-100m-round-and-nvidia-pact", "sourceType": "press", "accessed": "2026-05-09", "notes": "Independent coverage of Iambic Series B, NVIDIA collaboration, and AI-enabled discovery platform.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-causaly", "title": "Causaly official site", "publisher": "Causaly", "date": "2026", "url": "https://www.causaly.com/", "sourceType": "official", "accessed": "2026-05-09", "notes": "Primary source for Causaly agentic AI, knowledge graph, and life-sciences R&D workflows.", "source_tier": "B", "publisher_type": "company" }, { "id": "finsmes-causaly-series-b-202307", "title": "Causaly Raises $60M in Series B Funding", "publisher": "FinSMEs", "date": "2023-07-13", "url": "https://www.finsmes.com/2023/07/causaly-raises-60m-in-series-b-funding.html", "sourceType": "press", "accessed": "2026-05-09", "notes": "Independent funding source for Causaly Series B and AI-powered preclinical discovery positioning.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-iktos", "title": "Iktos official site", "publisher": "Iktos", "date": "2026", "url": "https://iktos.ai/", "sourceType": "official", "accessed": "2026-05-09", "notes": "Primary source for Iktos generative AI, robotics, and autonomous lab positioning.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-iktos-about", "title": "Iktos about", "publisher": "Iktos", "date": "2026", "url": "https://iktos.ai/about", "sourceType": "official", "accessed": "2026-05-09", "notes": "Company history source for Iktos founding year, platform milestones, Series A, and EIC grant.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-genesis-molecular-ai", "title": "Genesis Molecular AI official site", "publisher": "Genesis Molecular AI", "date": "2026", "url": "https://www.genesis.ml/", "sourceType": "official", "accessed": "2026-05-09", "notes": "Primary source for Genesis Molecular AI, GEMS, Pearl, and AI drug discovery platform.", "source_tier": "B", "publisher_type": "company" }, { "id": "fierce-genesis-series-b-202308", "title": "AI drug designer Genesis Therapeutics forges $200M round with eye on clinical testing", "publisher": "Fierce Biotech", "date": "2023-08-21", "url": "https://www.fiercebiotech.com/medtech/ai-drug-designer-genesis-therapeutics-forges-200m-round-eye-clinical-testing", "sourceType": "press", "accessed": "2026-05-09", "notes": "Independent source for Genesis Series B, GEMS platform, Stanford origin, and lab-in-the-loop validation.", "source_tier": "C", "publisher_type": "media" }, { "id": "cen-genesis-series-b-202308", "title": "Genesis Therapeutics raises $200 million for AI-aided drug discovery", "publisher": "Chemical & Engineering News", "date": "2023-08-24", "url": "https://cen.acs.org/business/informatics/Genesis-Therapeutics-raises-200-million/101/i28", "sourceType": "press", "accessed": "2026-05-09", "notes": "Chemistry-sector coverage of Genesis Series B, GEMS platform, and small-molecule AI discovery scope.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-argonne-autonomous-discovery", "title": "Autonomous Discovery", "publisher": "Argonne National Laboratory", "date": "2026", "url": "https://www.anl.gov/autonomous-discovery", "sourceType": "official", "accessed": "2026-05-09", "notes": "Primary source for Argonne Autonomous Discovery initiative, goals, projects, and autonomous-lab scope.", "source_tier": "B", "publisher_type": "academic_or_lab" }, { "id": "official-argonne-science101-autonomous-discovery", "title": "Science 101: Autonomous Discovery", "publisher": "Argonne National Laboratory", "date": "2026", "url": "https://www.anl.gov/science-101/autonomous-discovery", "sourceType": "official", "accessed": "2026-05-09", "notes": "Argonne explainer for self-driving labs, robotics, AI, and autonomous discovery examples.", "source_tier": "B", "publisher_type": "academic_or_lab" }, { "id": "official-ornl-autonomous-science", "title": "Autonomous Science", "publisher": "Oak Ridge National Laboratory", "date": "2026", "url": "https://www.ornl.gov/autonomousscience", "sourceType": "official", "accessed": "2026-05-09", "notes": "Primary source for ORNL autonomous laboratories, INTERSECT, Labs of the Future, and Genesis Mission role.", "source_tier": "B", "publisher_type": "academic_or_lab" }, { "id": "official-ornl-autonomous-chemistry-lab", "title": "An Autonomous Chemistry Lab for Accelerated Materials Discovery and Innovation", "publisher": "Oak Ridge National Laboratory", "date": "2026", "url": "https://www.ornl.gov/project/autonomous-chemistry-lab-accelerated-materials-discovery-and-innovation", "sourceType": "official", "accessed": "2026-05-09", "notes": "Project source for ORNL Autonomous Chemistry Lab robotics, synthesis, characterization, and AI integration.", "source_tier": "B", "publisher_type": "academic_or_lab" }, { "id": "official-llnl-apex-202507", "title": "Self-driving lab to automate the discovery of novel alloys", "publisher": "Lawrence Livermore National Laboratory", "date": "2025-07-21", "url": "https://www.llnl.gov/article/53131/self-driving-lab-automate-discovery-novel-alloys", "sourceType": "official", "accessed": "2026-05-09", "notes": "Primary source for LLNL APEX self-driving alloy platform and autonomous loop design.", "source_tier": "B", "publisher_type": "academic_or_lab" }, { "id": "str-llnl-apex-202601", "title": "Accelerating Alloy Development", "publisher": "Science & Technology Review / LLNL", "date": "2026-01", "url": "https://str.llnl.gov/str-januaryfebruary-2026/laboratory-news", "sourceType": "research", "accessed": "2026-05-09", "notes": "LLNL review summary confirming APEX goals, robotics, machine learning, and planned autonomous alloy workflow.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "id": "official-nist-autonomous-labs", "title": "Autonomous laboratories", "publisher": "NIST", "date": "2026", "url": "https://www.nist.gov/autonomous-laboratories", "sourceType": "official", "accessed": "2026-05-09", "notes": "NIST overview of autonomous laboratories, closed-loop AI, robotics, and biofoundry/materials scope.", "source_tier": "B", "publisher_type": "company" }, { "id": "nist-autonomous-systems-materials-202009", "title": "Autonomous Systems for Materials Research and Metrology", "publisher": "NIST", "date": "2020-08-31", "url": "https://www.nist.gov/programs-projects/autonomous-systems-materials-research-and-metrology-accelerating-discovery-and", "sourceType": "official", "accessed": "2026-05-09", "notes": "NIST project source for closed-loop autonomous systems, CAMEO, ANDiE, and materials-metrology applications.", "source_tier": "B", "publisher_type": "company" }, { "id": "nist-autonomous-formulation-lab-202508", "title": "Autonomous Formulation Lab", "publisher": "NIST", "date": "2025-08", "url": "https://www.nist.gov/programs-projects/autonomous-formulation-lab", "sourceType": "official", "accessed": "2026-05-09", "notes": "NIST source for Autonomous Formulation Lab, robotic sample prep, flow-cell measurements, and open-source platform.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-nrel-autonomous-experimentation", "title": "Autonomous Experimentation", "publisher": "NREL", "date": "2026", "url": "https://www.nrel.gov/materials-science/autonomous-experimentation", "sourceType": "official", "accessed": "2026-05-09", "notes": "Primary source for NREL autonomous synthesis, characterization, and AI-enhanced control software.", "source_tier": "B", "publisher_type": "academic_or_lab" }, { "id": "rsc-text-to-test-202410", "title": "From text to test: AI-generated control software for materials science instruments", "publisher": "Digital Discovery / Royal Society of Chemistry", "date": "2024-10-23", "url": "https://pubs.rsc.org/en/content/articlehtml/2024/dd/d4dd00143e", "sourceType": "research", "accessed": "2026-05-09", "notes": "Peer-reviewed NREL paper on LLM-generated instrument control software for autonomous experimentation.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "id": "official-bnl-nsls2-dssi", "title": "Data Science and Systems Integration Research and Development", "publisher": "Brookhaven National Laboratory / NSLS-II", "date": "2026", "url": "https://www.bnl.gov/nsls2/datascience/research-and-development.php", "sourceType": "official", "accessed": "2026-05-09", "notes": "BNL source for NSLS-II autonomous experiment steering, robotics, Bluesky, and AI decision-making work.", "source_tier": "B", "publisher_type": "academic_or_lab" }, { "id": "sciencedaily-bnl-ai-nanostructures-202301", "title": "AI discovers new nanostructures", "publisher": "ScienceDaily / Brookhaven National Laboratory", "date": "2023-01-13", "url": "https://www.sciencedaily.com/releases/2023/01/230113145356.htm", "sourceType": "press", "accessed": "2026-05-09", "notes": "Coverage of BNL autonomous methods discovering new nanostructures at the Center for Functional Nanomaterials.", "source_tier": "B", "publisher_type": "company_press_release" }, { "id": "official-mgi-about", "title": "About the Materials Genome Initiative", "publisher": "Materials Genome Initiative", "date": "2026", "url": "https://www.mgi.gov/about-materials-genome-initiative", "sourceType": "official", "accessed": "2026-05-09", "notes": "Primary source for MGI mission, 2011 origin, and accelerated materials development policy scope.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-mgi-autonomous-experimentation-report", "title": "Autonomous Experimentation for Materials R&D", "publisher": "Materials Genome Initiative", "date": "2026", "url": "https://www.mgi.gov/autonomous-experimentation-materials-rd", "sourceType": "official", "accessed": "2026-05-09", "notes": "MGI report page for autonomous experimentation and AMII workshop context.", "source_tier": "B", "publisher_type": "company" }, { "id": "sec-rxrx-2026q1", "title": "Recursion Pharmaceuticals Form 10-Q for the quarter ended March 31, 2026", "publisher": "SEC EDGAR", "date": "2026-05-06", "url": "https://www.sec.gov/Archives/edgar/data/1601830/000160183026000078/rxrx-20260331.htm", "sourceType": "regulatory", "accessed": "2026-05-10", "notes": "SEC filing and XBRL companyfacts source for Recursion Q1 2026 revenue, R&D expense, cash, operating cash flow, and liabilities.", "source_tier": "A", "publisher_type": "regulatory_registry" }, { "id": "sec-dna-2026q1", "title": "Ginkgo Bioworks Form 10-Q for the quarter ended March 31, 2026", "publisher": "SEC EDGAR", "date": "2026-05-07", "url": "https://www.sec.gov/Archives/edgar/data/1830214/000162828026032116/dna-20260331.htm", "sourceType": "regulatory", "accessed": "2026-05-13", "notes": "SEC filing and XBRL companyfacts source for Ginkgo Q1 2026 revenue, R&D expense, $143.9M cash and equivalents, $229.6M marketable securities, operating cash flow, and liabilities.", "source_tier": "A", "publisher_type": "regulatory_registry" }, { "id": "sec-abcl-2025-10k", "title": "AbCellera Form 10-K for the year ended December 31, 2025", "publisher": "SEC EDGAR", "date": "2026-02-24", "url": "https://www.sec.gov/Archives/edgar/data/1703057/000170305726000012/abcl-20251231.htm", "sourceType": "regulatory", "accessed": "2026-05-10", "notes": "SEC filing and XBRL companyfacts source for AbCellera FY2025 revenue, R&D expense, cash, operating cash flow, and lease liabilities.", "source_tier": "A", "publisher_type": "regulatory_registry" }, { "id": "clinicaltrials-nct05552755", "title": "Evaluate REC-4881 in Participants With Familial Adenomatous Polyposis (FAP)", "publisher": "ClinicalTrials.gov", "date": "2025-12-04", "url": "https://clinicaltrials.gov/study/NCT05552755", "sourceType": "regulatory", "accessed": "2026-05-10", "notes": "Clinical registry source for Recursion's TUPELO Phase 1b/2 REC-4881 trial in familial adenomatous polyposis.", "source_tier": "A", "publisher_type": "regulatory_registry" }, { "id": "nature-alab-202312", "title": "An autonomous laboratory for the accelerated synthesis of inorganic materials", "publisher": "Nature", "date": "2023-11-29", "url": "https://www.nature.com/articles/s41586-023-06734-w", "sourceType": "research", "accessed": "2026-05-10", "notes": "Peer-reviewed A-Lab paper later updated by a 2026 Nature Author Correction.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "id": "nature-alab-correction-202601", "title": "Author Correction: An autonomous laboratory for the accelerated synthesis of inorganic materials", "publisher": "Nature", "date": "2026-01-19", "url": "https://www.nature.com/articles/s41586-025-09992-y", "sourceType": "research", "accessed": "2026-05-10", "notes": "Nature correction clarifying A-Lab material-novelty claims and updating the reported successful-material count after post-publication re-analysis.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "id": "bloomberg-periodic-valuation-talks-202603", "title": "AI Science Startup Periodic Labs Is in Deal Talks at About $7 Billion Valuation", "publisher": "Bloomberg", "date": "2026-03-25", "url": "https://www.bloomberg.com/news/articles/2026-03-25/ai-science-startup-periodic-labs-is-in-deal-talks-at-about-7-billion-valuation", "sourceType": "press", "accessed": "2026-05-10", "notes": "Unnamed-source report for Periodic Labs valuation talks; kept out of funding totals as reported, not closed, valuation-only intelligence.", "source_tier": "C", "publisher_type": "media" }, { "id": "prnewswire-insilico-hkex-listing-202512", "title": "Insilico Medicine Lists on Hong Kong Stock Exchange, Showing AI Drug Discovery Momentum with 2025's Largest Hong Kong Biotech IPO", "publisher": "PR Newswire / Insilico Medicine", "date": "2025-12-30", "url": "https://www.prnewswire.com/news-releases/insilico-medicine-lists-on-hong-kong-stock-exchange-showing-ai-drug-discovery-momentum-with-2025s-largest-hong-kong-biotech-ipo-302650606.html", "sourceType": "press", "accessed": "2026-05-10", "notes": "Company release used for HKEX listing identity, stock code, and public-company status.", "source_tier": "B", "publisher_type": "company_press_release" }, { "id": "official-tecan-2025-results-202603", "title": "Tecan presents 2025 results and provides details on program to reignite profitable growth", "publisher": "Tecan", "date": "2026-03-16", "url": "https://www.tecan.com/corporate-news/tecan-presents-2025-results-and-provides-details-on-program-to-reignite-profitable-growth-63499", "sourceType": "official", "accessed": "2026-05-10", "notes": "Official issuer release used for SIX ticker, ISIN, public listing, and annual financial context pending normalization.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "nature-ai-scientist-202603", "title": "The AI Scientist: towards fully automated open-ended scientific discovery", "publisher": "Nature", "date": "2026-03-26", "url": "https://www.nature.com/articles/s41586-026-10265-5", "sourceType": "research", "notes": "Peer-reviewed Nature article for Sakana AI Scientist and AI Scientist-v2 autonomous research systems.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "arxiv-sakana-ai-scientist-202408", "title": "The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery", "publisher": "arXiv", "date": "2024-08-12", "url": "https://arxiv.org/abs/2408.06292", "sourceType": "research", "notes": "Original Sakana AI Scientist preprint describing autonomous ideation, experiment execution, paper writing, and review.", "source_tier": "B", "publisher_type": "preprint_server" }, { "accessed": "2026-05-11", "id": "arxiv-sakana-ai-scientist-v2-202504", "title": "The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search", "publisher": "arXiv", "date": "2025-04-10", "url": "https://arxiv.org/abs/2504.08066", "sourceType": "research", "notes": "Preprint reporting AI Scientist-v2, tree-search experimentation, and an AI-generated workshop paper passing peer review.", "source_tier": "B", "publisher_type": "preprint_server" }, { "accessed": "2026-05-11", "id": "github-sakana-ai-scientist-v2-202504", "title": "SakanaAI/AI-Scientist-v2", "publisher": "GitHub", "date": "2025-04-10", "url": "https://github.com/SakanaAI/AI-Scientist-v2", "sourceType": "repository", "notes": "Open-source AI Scientist-v2 codebase and README with implementation, risk, and citation details.", "source_tier": "B", "publisher_type": "code_repository" }, { "accessed": "2026-05-11", "id": "arxiv-paperqa2-202409", "title": "Language Agents Achieve Superhuman Synthesis of Scientific Knowledge", "publisher": "arXiv", "date": "2024-09-20", "url": "https://arxiv.org/abs/2409.13740", "sourceType": "research", "notes": "PaperQA2, WikiCrow, LitQA2, and ContraCrow preprint with exact LitQA2 and contradiction-detection measurements.", "source_tier": "B", "publisher_type": "preprint_server" }, { "accessed": "2026-05-11", "id": "nature-alphafold3-202405", "title": "Accurate structure prediction of biomolecular interactions with AlphaFold 3", "publisher": "Nature", "date": "2024-05-08", "url": "https://www.nature.com/articles/s41586-024-07487-w", "sourceType": "research", "notes": "Peer-reviewed AlphaFold 3 paper from Google DeepMind and Isomorphic Labs.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "nature-alphagenome-202506", "title": "Advancing regulatory variant effect prediction with AlphaGenome", "publisher": "Nature", "date": "2025-06-25", "url": "https://www.nature.com/articles/s41586-025-10014-0", "sourceType": "research", "notes": "Peer-reviewed AlphaGenome paper with genome-track and variant-effect benchmark aggregates.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "nature-gnome-202311", "title": "Scaling deep learning for materials discovery", "publisher": "Nature", "date": "2023-11-29", "url": "https://www.nature.com/articles/s41586-023-06735-9", "sourceType": "research", "notes": "Peer-reviewed GNoME paper reporting 2.2 million candidate structures and 381,000 newly stable crystals on the convex hull.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "nature-funsearch-202312", "title": "Mathematical discoveries from program search with large language models", "publisher": "Nature", "date": "2023-12-14", "url": "https://www.nature.com/articles/s41586-023-06924-6", "sourceType": "research", "notes": "Peer-reviewed FunSearch paper on LLM-guided program search for mathematical and algorithmic discovery.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "nature-gencast-202412", "title": "Probabilistic weather forecasting with machine learning", "publisher": "Nature", "date": "2024-12-04", "url": "https://www.nature.com/articles/s41586-024-08252-9", "sourceType": "research", "notes": "Peer-reviewed GenCast paper reporting probabilistic medium-range weather forecast performance.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "official-alphaevolve-202505", "title": "AlphaEvolve: A Gemini-powered coding agent for designing advanced algorithms", "publisher": "Google DeepMind", "date": "2025-05-14", "url": "https://deepmind.google/blog/alphaevolve-a-gemini-powered-coding-agent-for-designing-advanced-algorithms/", "sourceType": "official", "notes": "Official AlphaEvolve launch post reporting automated algorithm discovery and the 4x4 matrix multiplication improvement.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "arxiv-alphaevolve-202506", "title": "AlphaEvolve: A coding agent for scientific and algorithmic discovery", "publisher": "arXiv", "date": "2025-06-16", "url": "https://arxiv.org/abs/2506.13131", "sourceType": "research", "notes": "AlphaEvolve white paper with exact matrix multiplication tensor-rank comparisons and algorithmic discovery results.", "source_tier": "B", "publisher_type": "preprint_server" }, { "accessed": "2026-05-11", "id": "arxiv-alphaproteo-202409", "title": "De novo design of high-affinity protein binders with AlphaProteo", "publisher": "arXiv", "date": "2024-09-12", "url": "https://arxiv.org/abs/2409.08022", "sourceType": "research", "notes": "AlphaProteo technical report on de novo protein binder design and experimental success-rate claims.", "source_tier": "B", "publisher_type": "preprint_server" }, { "accessed": "2026-05-11", "id": "science-alphamissense-202309", "title": "Accurate proteome-wide missense variant effect prediction with AlphaMissense", "publisher": "Science", "date": "2023-09-19", "url": "https://www.science.org/doi/10.1126/science.adg7492", "sourceType": "research", "notes": "Peer-reviewed AlphaMissense paper; Google DeepMind publication page reports the Science venue and model scope.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "nature-foldbench-2025", "title": "FoldBench: systematic benchmark of all-atom biomolecular structure prediction models", "publisher": "Nature Communications", "date": "2025-12-01", "url": "https://www.nature.com/articles/s41467-025-67127-3", "sourceType": "research", "notes": "Independent FoldBench article with exact AlphaFold 3, Boltz-1, Chai-1, HelixFold 3, and Protenix measurements.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "official-recursion-openphenom-202411", "title": "Recursion Announces the Release of OpenPhenom-S/16 in Google Cloud Model Garden", "publisher": "Recursion Pharmaceuticals", "date": "2024-11-12", "url": "https://ir.recursion.com/news-releases/news-release-details/recursion-announces-release-openphenom-s16-google-clouds-model", "sourceType": "official", "notes": "Official release for OpenPhenom-S/16, a public phenomics foundation model for microscopy data.", "source_tier": "B", "publisher_type": "company_press_release" }, { "accessed": "2026-05-11", "id": "official-recursion-boltz2-202506", "title": "MIT and Recursion Release Boltz-2: Next Generation AI Model to Predict Binding Affinity", "publisher": "Recursion Pharmaceuticals", "date": "2025-06-06", "url": "https://ir.recursion.com/news-releases/news-release-details/mit-and-recursion-release-boltz-2-next-generation-ai-model/", "sourceType": "official", "notes": "Official MIT/Recursion Boltz-2 release, including open-source model and speed/accuracy positioning.", "source_tier": "B", "publisher_type": "company_press_release" }, { "accessed": "2026-05-11", "id": "official-recursion-brightfield-202409", "title": "How Phenomic Foundation Models Are Empowering the Brightfield Comeback", "publisher": "Recursion Pharmaceuticals", "date": "2024-09-11", "url": "https://www.recursion.com/news/how-phenomic-foundation-models-are-empowering-the-brightfield-comeback", "sourceType": "official", "notes": "Recursion technical blog describing Phenom brightfield transfer results and model capability claims.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "nature-generate-chroma-202311", "title": "Illuminating protein space with a programmable generative model", "publisher": "Nature", "date": "2023-11-15", "url": "https://www.nature.com/articles/s41586-023-06728-8", "sourceType": "research", "notes": "Peer-reviewed Chroma paper from Generate Biomedicines on programmable protein generation.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "official-generate-chroma-202311", "title": "Generate:Biomedicines Announces Publication of its Chroma Model in Nature", "publisher": "Generate Biomedicines", "date": "2023-11-15", "url": "https://generatebiomedicines.com/media-center/generatebiomedicines-announces-publication-of-its-chroma-model-in-nature", "sourceType": "official", "notes": "Company release for Chroma Nature publication and open-source availability.", "source_tier": "B", "publisher_type": "company_press_release" }, { "accessed": "2026-05-11", "id": "official-iambic-neuralplexer-202402", "title": "Iambic announces Nature Machine Intelligence cover article for NeuralPLexer", "publisher": "Iambic Therapeutics", "date": "2024-02-12", "url": "https://www.iambic.ai/post/iambic-therapeutics-announces-new-research-with-cover-article-in-nature-machine-intelligence-demonstrating-the-capabilities-of-its-generative-ai-neuralplexer-technology-to-predict-protein-ligand-complex-structures", "sourceType": "official", "notes": "Company release for NeuralPLexer protein-ligand structure prediction research in Nature Machine Intelligence.", "source_tier": "B", "publisher_type": "company_press_release" }, { "accessed": "2026-05-11", "id": "nature-comm-insitro-posh-202512", "title": "A pooled Cell Painting CRISPR screening platform enables de novo inference of gene function by self-supervised deep learning", "publisher": "Nature Communications", "date": "2025-12-16", "url": "https://www.nature.com/articles/s41467-025-66778-6", "sourceType": "research", "notes": "Peer-reviewed insitro POSH paper with code and model availability statements.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "official-insitro-posh-202512", "title": "insitro Validates AI-Enabled POSH Platform in Nature Communications", "publisher": "insitro", "date": "2025-12-16", "url": "https://www.insitro.com/news/insitro-validates-ai-enabled-posh-platform-in-nature-communications-bridging-critical-gap-in-drug-discovery/", "sourceType": "official", "notes": "Company release contextualizing the POSH Nature Communications publication.", "source_tier": "B", "publisher_type": "company_press_release" }, { "accessed": "2026-05-11", "id": "arxiv-labbench-202407", "title": "LAB-Bench: Measuring Capabilities of Language Models for Biology Research", "publisher": "arXiv", "date": "2024-07-13", "url": "https://arxiv.org/abs/2407.10362", "sourceType": "research", "notes": "LAB-Bench preprint with exact biology research task measurements for humans and frontier models.", "source_tier": "B", "publisher_type": "preprint_server" }, { "accessed": "2026-05-11", "id": "official-autoscience-carl", "title": "Carl: Autonomous AI Research Agent", "publisher": "Autoscience Institute", "date": "2026", "url": "https://www.autoscience.ai/carl", "sourceType": "official", "notes": "Official Carl product page describing autonomous ideation, experimentation, paper-writing, and peer-review submission workflow.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-autoscience-mira", "title": "Mira: Productionizing Research", "publisher": "Autoscience Institute", "date": "2026", "url": "https://www.autoscience.ai/mira", "sourceType": "official", "notes": "Official Mira page describing conversion of research outputs into production ML model changes.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-chemify-technology", "title": "Chemify technology platform", "publisher": "Chemify", "date": "2026", "url": "https://chemify.io/technology", "sourceType": "official", "notes": "Official technology page for Chemify molecular design, chemputation, and automated synthesis capabilities.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "nature-chemputer-202012", "title": "A universal system for digitization and automatic execution of the chemical synthesis literature", "publisher": "Science", "date": "2020-12-04", "url": "https://www.science.org/doi/10.1126/science.abc2986", "sourceType": "research", "notes": "Peer-reviewed Chemputer paper from the Cronin group on automatic execution of chemical synthesis procedures.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "nature-comm-chemputer-engine-202402", "title": "Chemputation and the Standardization of Chemical Informatics", "publisher": "Nature Communications", "date": "2024-02-01", "url": "https://www.nature.com/articles/s41467-024-44873-6", "sourceType": "research", "notes": "Peer-reviewed paper on chemputation architecture and standardization underlying digital chemistry workflows.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "nature-comm-chemputer-blueprints-202411", "title": "Automatic conversion of chemical synthesis procedures into digital executable blueprints", "publisher": "Nature Communications", "date": "2024-11-14", "url": "https://www.nature.com/articles/s41467-024-53921-x", "sourceType": "research", "notes": "Peer-reviewed paper on converting synthesis procedures into executable chemical blueprints.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "comms-chem-chemputation-llm-202604", "title": "Chemputation with large language models", "publisher": "Communications Chemistry", "date": "2026-04-14", "url": "https://www.nature.com/articles/s42004-026-01510-9", "sourceType": "research", "notes": "Peer-reviewed article on using LLMs for chemputation workflows.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "rsc-ecl-hplc-anomaly-202510", "title": "Self-supervised anomaly detection for HPLC traces in a cloud laboratory", "publisher": "Digital Discovery", "date": "2025-10-08", "url": "https://pubs.rsc.org/en/content/articlelanding/2025/dd/d5dd00153a", "sourceType": "research", "notes": "Peer-reviewed Digital Discovery paper using Emerald Cloud Lab generated HPLC traces for anomaly detection.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "official-highres-cellario", "title": "Cellario laboratory automation scheduler", "publisher": "HighRes Biosolutions", "date": "2026", "url": "https://www.highres.com/lab-orchestration", "sourceType": "official", "notes": "Official Cellario lab orchestration page for scheduling, simulation, and execution of automated laboratory workflows.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-highres-tetrascience-integration", "title": "Cellario and Tetra Data Platform cloud integration", "publisher": "HighRes Biosolutions", "date": "2026", "url": "https://www.highres.com/highres-blog/cloud-based-management-with-lab-automation", "sourceType": "official", "notes": "HighRes page describing integration between Cellario whole-lab automation software and TetraScience Tetra Data Platform.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-automata-linq", "title": "LINQ cloud laboratory automation platform", "publisher": "Automata", "date": "2026", "url": "https://www.automata.tech/linq", "sourceType": "official", "notes": "Official LINQ platform page for Automata robotic lab automation and orchestration.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "docs-opentrons-flex", "title": "Opentrons Flex robot documentation", "publisher": "Opentrons", "date": "2026", "url": "https://docs.opentrons.com/flex/", "sourceType": "official", "notes": "Official Opentrons Flex documentation for liquid-handling and lab automation capability surface.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-tecan-fluent", "title": "Fluent automated workstation", "publisher": "Tecan", "date": "2026", "url": "https://www.tecan.com/fluent-automated-workstation", "sourceType": "official", "notes": "Official Tecan Fluent product page covering automated liquid handling and assay automation.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-tecan-fluentcontrol", "title": "Fluent automated workstation and control software", "publisher": "Tecan", "date": "2026", "url": "https://www.tecan.com/fluent-automated-workstation", "sourceType": "official", "notes": "Official Tecan Fluent page used as public technical source for workstation and workflow-control capability.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-tetrascience-tetra-data", "title": "Tetra Scientific Data and AI Platform", "publisher": "TetraScience", "date": "2026", "url": "https://www.tetrascience.com/platform/tetra-scientific-data-ai-cloud", "sourceType": "official", "notes": "Official Tetra Scientific Data and AI Platform page for vendor-neutral scientific data engineering and AI-ready data products.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-tetrascience-tetra-os", "title": "Tetra Scientific Data and AI Platform", "publisher": "TetraScience", "date": "2026", "url": "https://www.tetrascience.com/platform/tetra-scientific-data-ai-cloud", "sourceType": "official", "notes": "Official Tetra platform page used for scientific data cloud infrastructure and AI-ready data claims.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-synthace-how-it-works", "title": "How Synthace works", "publisher": "Synthace", "date": "2026", "url": "https://www.synthace.com/how-it-works", "sourceType": "official", "notes": "Official Synthace page describing experiment design, automation, and data capture workflows.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-causaly-knowledge-graph", "title": "Causaly Knowledge Graph", "publisher": "Causaly", "date": "2026", "url": "https://www.causaly.com/life-science-ai/knowledge-graph", "sourceType": "official", "notes": "Official Causaly Knowledge Graph page describing biomedical evidence extraction, graph scale, and scientific RAG.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-causaly-bio-graph", "title": "Bio Graph product", "publisher": "Causaly", "date": "2026", "url": "https://www.causaly.com/products/bio-graph", "sourceType": "official", "notes": "Official Bio Graph product page for drug discovery and biomedical relationship exploration.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-atinary-semopt", "title": "SDLabs and no-code self-driving lab optimization", "publisher": "Atinary", "date": "2026", "url": "https://atinary.com/sdlabs/", "sourceType": "official", "notes": "Official Atinary SDLabs page for no-code self-driving lab optimization workflows.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "chemrxiv-semopt-202205", "title": "Bayesian optimization for self-driving labs with unknown constraints", "publisher": "ChemRxiv", "date": "2022-05-09", "url": "https://chemrxiv.org/engage/chemrxiv/article-details/6279374d29bae9a261ecdd6d", "sourceType": "research", "notes": "Atinary-associated ChemRxiv preprint on constraint-aware Bayesian optimization for self-driving laboratories.", "source_tier": "B", "publisher_type": "preprint_server" }, { "accessed": "2026-05-11", "id": "citrine-materials-data-facility-2016", "title": "The Citrination platform: a materials data facility", "publisher": "JOM", "date": "2016-09-27", "url": "https://link.springer.com/article/10.1007/s11837-016-2102-7", "sourceType": "research", "notes": "Peer-reviewed JOM article describing Citrine Informatics Citrination materials data platform.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "official-pnnl-autonomous-science", "title": "Autonomous science and self-driving laboratories", "publisher": "PNNL", "date": "2026", "url": "https://www.pnnl.gov/autonomous-science", "sourceType": "official", "notes": "PNNL autonomous science program page covering AI-guided experiments and lab automation.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-pnnl-energy-storage-hte", "title": "High-throughput experimentation for energy storage materials", "publisher": "PNNL", "date": "2026", "url": "https://www.pnnl.gov/projects/high-throughput-experimentation-energy-storage-materials", "sourceType": "official", "notes": "PNNL project page for autonomous and high-throughput experimentation in energy storage materials.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "nature-comm-self-driving-labs-202504", "title": "Acceleration of scientific discovery by self-driving laboratories", "publisher": "Nature Communications", "date": "2025-04-04", "url": "https://www.nature.com/articles/s41467-025-57940-2", "sourceType": "research", "notes": "Peer-reviewed Acceleration Consortium paper on self-driving laboratories and accelerated discovery.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "nature-synthesis-rise-sdl-202301", "title": "The rise of self-driving labs in chemical and materials sciences", "publisher": "Nature Synthesis", "date": "2023-01-12", "url": "https://www.nature.com/articles/s44160-022-00231-0", "sourceType": "research", "notes": "Review article on self-driving laboratories in chemistry and materials science.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "official-bnl-ai-nanostructures-202301", "title": "AI discovers new nanostructures", "publisher": "Brookhaven National Laboratory", "date": "2023-01-12", "url": "https://www.bnl.gov/newsroom/news.php?a=120978", "sourceType": "official", "notes": "BNL release on autonomous AI-guided X-ray scattering analysis for nanostructure discovery.", "source_tier": "B", "publisher_type": "government_lab" }, { "accessed": "2026-05-11", "id": "science-advances-bnl-nanostructures-202301", "title": "Autonomous materials discovery by X-ray scattering", "publisher": "Science Advances", "date": "2023-01-11", "url": "https://www.science.org/doi/10.1126/sciadv.add3687", "sourceType": "research", "notes": "Peer-reviewed Science Advances paper associated with BNL AI-guided nanostructure discovery.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "nist-composable-lab-ecosystem-202605", "title": "A composable laboratory ecosystem for autonomous experimentation", "publisher": "NIST", "date": "2026-05-07", "url": "https://www.nist.gov/publications/composable-laboratory-ecosystem-autonomous-experimentation", "sourceType": "research", "notes": "NIST publication page for composable autonomous laboratory ecosystem work.", "source_tier": "A", "publisher_type": "government_lab" }, { "accessed": "2026-05-11", "id": "official-generate-phase1-gb0895-202509", "title": "Generate:Biomedicines to present Phase 1 results for GB-0895", "publisher": "Generate Biomedicines", "date": "2025-09-25", "url": "https://generatebiomedicines.com/media-center/generatebiomedicines-to-present-phase-1-results", "sourceType": "official", "notes": "Company release for Phase 1 GB-0895 results presentation at ERS 2025.", "source_tier": "B", "publisher_type": "company_press_release" }, { "accessed": "2026-05-11", "id": "official-generate-q1-202605", "title": "Generate Biomedicines reports Q1 2026 financial results and business update", "publisher": "Generate Biomedicines / PRNewswire", "date": "2026-05-07", "url": "https://www.prnewswire.com/news-releases/generate-biomedicines-inc-reports-first-quarter-2026-financial-results-and-provides-business-update-302765449.html", "sourceType": "official", "notes": "PRNewswire-hosted company release describing Q1 2026 pipeline progress and SOLAIRIA trials.", "source_tier": "B", "publisher_type": "company_press_release" }, { "accessed": "2026-05-11", "id": "clinicaltrials-nct07359846", "title": "GB-0895 in adults and adolescents with severe asthma", "publisher": "ClinicalTrials.gov", "date": "2026-01-20", "url": "https://clinicaltrials.gov/study/NCT07359846", "sourceType": "regulatory", "notes": "ClinicalTrials.gov registry entry for SOLAIRIA-1 GB-0895 severe asthma trial.", "source_tier": "A", "publisher_type": "regulatory_database" }, { "accessed": "2026-05-11", "id": "clinicaltrials-nct06253871", "title": "IAM1363 in HER2-altered cancers", "publisher": "ClinicalTrials.gov", "date": "2024-03-25", "url": "https://clinicaltrials.gov/study/NCT06253871", "sourceType": "regulatory", "notes": "ClinicalTrials.gov registry entry for Iambic IAM1363 Phase 1/1b HER2-altered cancer trial.", "source_tier": "A", "publisher_type": "regulatory_database" }, { "accessed": "2026-05-11", "id": "official-iambic-iam1363-first-patient-202403", "title": "Iambic doses first patient in IAM1363 trial", "publisher": "Iambic Therapeutics", "date": "2024-03-28", "url": "https://www.iambic.ai/post/iambic-therapeutics-doses-first-patient-in-phase-1-clinical-study-of-iam1363", "sourceType": "official", "notes": "Company release for first patient dosed in IAM1363 Phase 1 study.", "source_tier": "B", "publisher_type": "company_press_release" }, { "accessed": "2026-05-11", "id": "clinicaltrials-nct05938920", "title": "INS018_055 in idiopathic pulmonary fibrosis", "publisher": "ClinicalTrials.gov", "date": "2023-07-07", "url": "https://clinicaltrials.gov/study/NCT05938920", "sourceType": "regulatory", "notes": "ClinicalTrials.gov registry entry for Insilico ISM001-055 / rentosertib idiopathic pulmonary fibrosis study.", "source_tier": "A", "publisher_type": "regulatory_database" }, { "accessed": "2026-05-11", "id": "nature-med-insilico-rentosertib-202506", "title": "A generative AI-discovered TNIK inhibitor for idiopathic pulmonary fibrosis", "publisher": "Nature Medicine", "date": "2025-06-01", "url": "https://www.nature.com/articles/s41591-025-03770-1", "sourceType": "research", "notes": "Peer-reviewed Nature Medicine clinical-stage report on AI-discovered TNIK inhibitor rentosertib for IPF.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "official-insilico-rentosertib-natmed-202506", "title": "Rentosertib asset profile", "publisher": "Insilico Medicine", "date": "2026", "url": "https://insilico.com/interactive/drugs/rentosertib", "sourceType": "official", "notes": "Insilico asset profile for rentosertib with Phase IIa status and clinical data summary.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "clinicaltrials-nct04411628", "title": "LY-CoV555 in participants hospitalized with COVID-19", "publisher": "ClinicalTrials.gov", "date": "2020-06-08", "url": "https://clinicaltrials.gov/study/NCT04411628", "sourceType": "regulatory", "notes": "ClinicalTrials.gov registry entry for Lilly/AbCellera LY-CoV555 COVID-19 antibody study.", "source_tier": "A", "publisher_type": "regulatory_database" }, { "accessed": "2026-05-11", "id": "official-abcellera-lilly-covid-202003", "title": "AbCellera and Lilly to co-develop antibody therapies for COVID-19", "publisher": "AbCellera", "date": "2020-03-12", "url": "https://investors.abcellera.com/news/news-releases/2020/AbCellera-and-Lilly-to-Co-develop-Antibody-Therapies-for-the-Treatment-of-COVID-19/default.aspx", "sourceType": "official", "notes": "AbCellera investor release announcing COVID-19 antibody collaboration with Lilly.", "source_tier": "B", "publisher_type": "company_press_release" }, { "accessed": "2026-05-11", "id": "official-ginkgo-automation-platform", "title": "Ginkgo Automation platform", "publisher": "Ginkgo Bioworks", "date": "2026", "url": "https://www.ginkgobioworks.com/automation/", "sourceType": "official", "notes": "Official page describing Ginkgo Automation robotic foundry and data-generation platform.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "nature-ginkgo-operating-system-201909", "title": "A programming language for biological engineering", "publisher": "Nature Biotechnology", "date": "2019-09-23", "url": "https://www.nature.com/articles/s41587-019-0249-7", "sourceType": "research", "notes": "Peer-reviewed Nature Biotechnology paper from Ginkgo-associated authors on programming biological engineering workflows.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "official-cellarity-science-framework-202510", "title": "Cellarity publishes framework for cell state-correcting medicines", "publisher": "Cellarity", "date": "2025-10-02", "url": "https://cellarity.com/news_item/cellarity-publishes-in-science-a-framework-for-the-discovery-of-cell-state-correcting-medicines/", "sourceType": "official", "notes": "Company release for Science paper on Cellarity cell-state correcting medicine framework.", "source_tier": "B", "publisher_type": "company_press_release" }, { "accessed": "2026-05-11", "id": "science-cellarity-framework-202510", "title": "A framework for the discovery of cell state-correcting medicines", "publisher": "Science", "date": "2025-10-03", "url": "https://www.science.org/doi/10.1126/science.adu7875", "sourceType": "research", "notes": "Peer-reviewed Science article describing Cellarity platform framework for cell-state correcting medicines.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "cellarity-open-single-cell-dataset-202208", "title": "Cellarity releases open single-cell dataset", "publisher": "Cellarity", "date": "2022-08-23", "url": "https://cellarity.com/news_item/cellarity-releases-open-single-cell-dataset-to-advance-cell-based-drug-discovery/", "sourceType": "official", "notes": "Company release describing open single-cell dataset for cell-based drug discovery.", "source_tier": "B", "publisher_type": "company_press_release" }, { "accessed": "2026-05-11", "id": "official-cellarity-cly124-phase1-202506", "title": "Cellarity advances CLY-124 into Phase 1 trial", "publisher": "Cellarity", "date": "2025-06-12", "url": "https://cellarity.com/news_item/cellarity-announces-fda-clearance-of-ind-application-for-cly-124/", "sourceType": "official", "notes": "Company release on FDA clearance and Phase 1 transition for CLY-124.", "source_tier": "B", "publisher_type": "company_press_release" }, { "accessed": "2026-05-11", "id": "official-cellarity-novo-mash-202401", "title": "Cellarity and Novo Nordisk collaborate on MASH", "publisher": "Cellarity", "date": "2024-01-04", "url": "https://cellarity.com/news_item/cellarity-announces-collaboration-with-novo-nordisk-to-develop-novel-therapies-for-mash/", "sourceType": "official", "notes": "Company collaboration release with Novo Nordisk for MASH therapies.", "source_tier": "B", "publisher_type": "company_press_release" }, { "accessed": "2026-05-11", "id": "official-helix-virtual-lab", "title": "Helical virtual AI lab", "publisher": "Helical", "date": "2026", "url": "https://www.helical.bio/", "sourceType": "official", "notes": "Official Helical site describing virtual AI lab for reproducible in-silico discovery.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-fathom-platform", "title": "Fathom Therapeutics physics and AI platform", "publisher": "Fathom Therapeutics", "date": "2026", "url": "https://www.fathomtx.com/", "sourceType": "official", "notes": "Official Fathom site describing physics- and AI-enabled small-molecule design platform.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-10x-science-platform", "title": "10x Science protein characterization platform", "publisher": "10x Science", "date": "2026", "url": "https://10xscience.com/", "sourceType": "official", "notes": "Official 10x Science site describing next-generation protein characterization platform.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-genesis-platform", "title": "Genesis AI Research", "publisher": "Genesis Molecular AI", "date": "2026", "url": "https://www.genesis.ml/ai-research", "sourceType": "official", "notes": "Official Genesis AI Research page describing foundation models, Pearl, reinforcement learning, multimodal learning, and agentic workflows for molecular design.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-xaira-platform", "title": "Xaira Therapeutics approach", "publisher": "Xaira Therapeutics", "date": "2026", "url": "https://www.xaira.com/our-approach", "sourceType": "official", "notes": "Official Xaira approach page covering advanced AI research, data generation, and therapeutic product development.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-ami-labs-platform", "title": "AMI Labs platform", "publisher": "AMI Labs", "date": "2026", "url": "https://amilabs.xyz/", "sourceType": "official", "notes": "Official AMI Labs site describing AI-native experimental workflows.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-chemlex-platform", "title": "ChemLex approach", "publisher": "ChemLex", "date": "2026", "url": "https://www.chemlex.com/our-approach.html", "sourceType": "official", "notes": "Official ChemLex approach page describing closed-loop dry lab / wet lab platform, route design, LCMS interpretation, and high-throughput automated synthesis.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-radical-ai-platform", "title": "Radical AI platform", "publisher": "Radical AI", "date": "2026", "url": "https://radicalai.com/", "sourceType": "official", "notes": "Official Radical AI site describing AI scientist workflow capabilities.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-thesis-platform", "title": "Thesis company profile", "publisher": "Y Combinator", "date": "2025", "url": "https://www.ycombinator.com/companies/thesis", "sourceType": "official", "notes": "Y Combinator profile for Thesis AI science tooling.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-zeon-platform", "title": "Zeon Systems platform", "publisher": "Zeon Systems", "date": "2026", "url": "https://zeonsystems.ai/", "sourceType": "official", "notes": "Official Zeon Systems site describing AI science workflow platform.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "jcheminf-iktos-synthetic-accessibility-202309", "title": "Integrating synthetic accessibility with AI-based generative drug design", "publisher": "Journal of Cheminformatics", "date": "2023-09-19", "url": "https://jcheminf.biomedcentral.com/articles/10.1186/s13321-023-00742-8", "sourceType": "research", "notes": "Peer-reviewed Iktos-authored paper integrating synthetic accessibility into generative drug design workflows.", "source_tier": "A", "publisher_type": "peer_reviewed_journal" }, { "accessed": "2026-05-11", "id": "official-iktos-makya", "title": "Makya generative AI platform", "publisher": "Iktos", "date": "2026", "url": "https://iktos.ai/solution/makya", "sourceType": "official", "notes": "Official Makya page describing Iktos generative AI platform for de novo molecule design.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-iktos-publications-go-lo-202509", "title": "Growing and Linking Optimizers: Synthesis-driven Molecule Design", "publisher": "Iktos", "date": "2025-09-25", "url": "https://iktos.ai/resources/publications/growing-and-linking-optimizers-synthesis-driven-molecule-design", "sourceType": "official", "notes": "Iktos publication page for reaction-based generative models GO and LO.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-causaly-technology", "title": "Causaly Bio Graph platform", "publisher": "Causaly", "date": "2026", "url": "https://www.causaly.com/products/bio-graph", "sourceType": "official", "notes": "Official Causaly product page describing evidence graph, biomedical search, and AI-assisted discovery workflow.", "source_tier": "B", "publisher_type": "company" }, { "accessed": "2026-05-11", "id": "official-generate-amgen-202201", "title": "Amgen and Generate Biomedicines announce multi-target, multi-modality research collaboration agreement", "publisher": "Amgen / Generate Biomedicines", "date": "2022-01-06", "url": "https://www.amgen.com/newsroom/press-releases/2022/01/amgen-and-generate-biomedicines-announce-multitarget-multimodality-research-collaboration-agreement", "sourceType": "official", "notes": "Primary collaboration release reporting $50M upfront and $1.9B potential transaction value.", "source_tier": "B", "publisher_type": "company_press_release" }, { "accessed": "2026-05-11", "id": "official-insitro-bms-202010", "title": "insitro announces five-year discovery collaboration with Bristol Myers Squibb", "publisher": "insitro", "date": "2020-10-28", "url": "https://www.insitro.com/news/insitro-announces-five-year-discovery-collaboration-with-bristol-myers-squibb-to-discover-and-develop-novel-treatments-for-amyotrophic-lateral-sclerosis-and-frontotemporal-dementia-2/", "sourceType": "official", "notes": "Primary collaboration release reporting $50M upfront and over $2.1B total potential deal value.", "source_tier": "B", "publisher_type": "company_press_release" }, { "accessed": "2026-05-11", "id": "official-insitro-bms-extension-202510", "title": "insitro extends research collaboration with Bristol Myers Squibb", "publisher": "insitro", "date": "2025-10-14", "url": "https://www.insitro.com/news/insitro-extends-research-collaboration-with-bristol-myers-squibb-leveraging-insitros-chemml-discovery-platform/", "sourceType": "official", "notes": "Collaboration extension release for insitro-BMS ChemML discovery work.", "source_tier": "B", "publisher_type": "company_press_release" }, { "id": "official-slas-2026", "title": "SLAS2026 International Conference and Exhibition", "publisher": "SLAS", "date": "2026-02", "url": "https://slas2026.org/", "sourceType": "official", "accessed": "2026-05-13", "notes": "Official SLAS 2026 event site for the Boston conference and exhibition, February 7-11, 2026.", "source_tier": "B", "publisher_type": "conference" }, { "id": "official-cenevo-rebrand-202507", "title": "Labguru and Titian Software rebrand as Cenevo", "publisher": "Cenevo", "date": "2025-07-30", "url": "https://www.cenevo.com/press/labguru-and-titian-software-rebrand-as-cenevo", "sourceType": "official", "accessed": "2026-05-13", "notes": "Cenevo rebrand announcement for Labguru and Titian Software, reporting 950+ customer organizations, 45,000+ users, and use by 8 of the top 10 pharma companies.", "source_tier": "B", "publisher_type": "company" }, { "id": "rdworld-cenevo-ai-agents-202602", "title": "Cenevo launches two AI agents for lab protocol conversion and workflow automation", "publisher": "R&D World", "date": "2026-02-10", "url": "https://www.rdworldonline.com/cenevo-launches-two-ai-agents-for-lab-protocol-conversion-and-workflow-automation/", "sourceType": "press", "accessed": "2026-05-13", "notes": "Trade coverage of Cenevo's AI Protocol Conversion and AI Automation agents, including workflow generation, human approvals, audit trails, and regulated-lab positioning.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-abb-slas-2026", "title": "ABB at SLAS 2026", "publisher": "ABB Robotics", "date": "2026-02", "url": "https://www.abb.com/global/en/areas/robotics/events/slas-2026", "sourceType": "official", "accessed": "2026-05-13", "notes": "ABB Robotics SLAS 2026 event page describing GoFa collaborative robot workcells and interoperability demonstrations with laboratory instrumentation partners.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-biosero-gosimple-202602", "title": "GoSimple", "publisher": "Biosero", "date": "2026", "url": "https://biosero.com/gosimple/", "sourceType": "official", "accessed": "2026-05-13", "notes": "Biosero product page for GoSimple standardized, pre-validated automation workcells.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-biosero-green-button-go-scheduler", "title": "Green Button Go Scheduler", "publisher": "Biosero", "date": "2026", "url": "https://biosero.com/products/software/green-button-go-scheduler/", "sourceType": "official", "accessed": "2026-05-13", "notes": "Biosero official page for Green Button Go Scheduler, the scheduling software layer used in Biosero automation workcells.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-chai-discovery", "title": "Chai Discovery official site", "publisher": "Chai Discovery", "date": "2026", "url": "https://www.chaidiscovery.com/", "sourceType": "official", "accessed": "2026-05-13", "notes": "Official Chai Discovery site for company identity, product positioning, and molecular-design platform scope.", "source_tier": "B", "publisher_type": "company" }, { "id": "businesswire-chai-series-b-202512", "title": "Chai Discovery Announces $130 Million Series B To Transform Molecular Discovery", "publisher": "Chai Discovery via Business Wire", "date": "2025-12-15", "url": "https://www.businesswire.com/news/home/20251214931432/en/Chai-Discovery-Announces-%24130-Million-Series-B-To-Transform-Molecular-Discovery", "sourceType": "official", "accessed": "2026-05-13", "notes": "Primary financing release reporting $130M Series B, $1.3B valuation, more than $225M total funding, Chai-2 zero-shot antibody design, and 100-fold improvement over previous computational methods.", "source_tier": "B", "publisher_type": "company_press_release" }, { "id": "techcrunch-chai-series-b-202512", "title": "OpenAI-backed biotech firm Chai Discovery raises $130M Series B at $1.3B valuation", "publisher": "TechCrunch", "date": "2025-12-15", "url": "https://techcrunch.com/2025/12/15/openai-backed-biotech-firm-chai-discovery-raises-130m-series-b-at-1-3b-valuation/", "sourceType": "press", "accessed": "2026-05-13", "notes": "Independent financing coverage confirming the Series B amount, valuation, investor syndicate, and total funding above $225M.", "source_tier": "C", "publisher_type": "media" }, { "id": "businesswire-chai2-202506", "title": "Chai Discovery Unveils Chai-2 Breakthrough, Achieving Fully De Novo Antibody Design With AI", "publisher": "Chai Discovery via Business Wire", "date": "2025-06-30", "url": "https://www.businesswire.com/news/home/20250630307418/en/Chai-Discovery-Unveils-Chai-2-Breakthrough-Achieving-Fully-De-Novo-Antibody-Design-With-AI", "sourceType": "official", "accessed": "2026-05-13", "notes": "Primary Chai-2 announcement reporting double-digit antibody hit rates, fewer than 20 experimental designs per target, and hit rates close to 20% versus traditional methods below 0.1%.", "source_tier": "B", "publisher_type": "company_press_release" }, { "id": "biorxiv-chai2-zero-shot-202507", "title": "Zero-shot antibody design in a 24-well plate", "publisher": "bioRxiv", "date": "2025-07-06", "url": "https://www.biorxiv.org/content/10.1101/2025.07.05.663018v1", "sourceType": "research", "accessed": "2026-05-13", "notes": "Chai-2 preprint reporting a 16% hit rate in de novo antibody design, over 100-fold improvement over previous computational methods, and a 68% wet-lab success rate for miniprotein design.", "source_tier": "B", "publisher_type": "preprint_server" }, { "id": "biospace-chai-lilly-202601", "title": "Chai Discovery Announces Collaboration with Eli Lilly and Company to Accelerate Biologics Discovery", "publisher": "Chai Discovery via BioSpace", "date": "2026-01-12", "url": "https://www.biospace.com/press-releases/chai-discovery-announces-collaboration-with-eli-lilly-and-company-to-accelerate-biologics-discovery", "sourceType": "official", "accessed": "2026-05-13", "notes": "Press release describing Lilly deployment of Chai's frontier AI platform and a purpose-built model trained on proprietary Lilly data for biologics discovery workflows.", "source_tier": "B", "publisher_type": "company_press_release" }, { "id": "official-astabench", "title": "AstaBench: Benchmarking AI Agents for Science", "publisher": "Ai2", "date": "2026", "url": "https://allenai.org/asta/bench", "sourceType": "official", "accessed": "2026-05-13", "notes": "Official AstaBench suite and leaderboard page for scientific-agent evaluations across literature understanding, coding/execution, data analysis, and end-to-end discovery tasks.", "source_tier": "B", "publisher_type": "nonprofit_research" }, { "id": "ai2-astabench-blog-202508", "title": "AstaBench: Rigorous benchmarking of AI agents with a holistic scientific research suite", "publisher": "Ai2", "date": "2025-08-26", "url": "https://allenai.org/blog/astabench", "sourceType": "research", "accessed": "2026-05-13", "notes": "Ai2 launch blog reporting 2,400+ scientific research problems, four task areas, standardized tools, agent-eval infrastructure, and early 53.0% Asta v0 overall result.", "source_tier": "B", "publisher_type": "nonprofit_research" }, { "id": "ai2-astabench-update-202604", "title": "AstaBench update: New results, plus adoption from industry", "publisher": "Ai2", "date": "2026-04-30", "url": "https://allenai.org/blog/astabench-update-spring-2026", "sourceType": "research", "accessed": "2026-05-13", "notes": "Ai2 spring 2026 update reporting new frontier-model AstaBench results including Claude Opus 4.7 at 58.0%, GPT-5.5 at 52.9%, and GPT-5.4 at 46.5%.", "source_tier": "B", "publisher_type": "nonprofit_research" }, { "id": "nature-robochem-flex-202604", "title": "A flexible and affordable self-driving laboratory for automated reaction optimization", "publisher": "Nature Synthesis", "date": "2026-04-13", "url": "https://www.nature.com/articles/s44160-026-01053-0", "sourceType": "research", "accessed": "2026-05-13", "notes": "Open-access Nature Synthesis paper introducing RoboChem-Flex, a low-cost modular self-driving laboratory with Python software, Bayesian optimization, closed-loop and human-in-the-loop modes, and six validation case studies.", "source_tier": "A", "publisher_type": "journal" }, { "id": "official-converge-series-a-202601", "title": "Converge Bio raises $25M in Series A to accelerate drug discovery with generative AI", "publisher": "Converge Bio", "date": "2026-01-13", "url": "https://converge-bio.com/news/converge-bio-raises-25m-in-series-a-to-accelerate-drug-discovery-with-generative-ai", "sourceType": "official", "accessed": "2026-05-13", "notes": "Primary Converge Bio release reporting $25M Series A, $30M total raised, Bessemer lead, over a dozen customers, 40+ programs, and use cases across target discovery, antibody design, and protein manufacturing optimization.", "source_tier": "B", "publisher_type": "company_press_release" }, { "id": "techcrunch-converge-series-a-202601", "title": "Converge Bio raises $25M, backed by Bessemer and execs from Meta, OpenAI, Wiz", "publisher": "TechCrunch", "date": "2026-01-13", "url": "https://techcrunch.com/2026/01/13/ai-drug-discovery-startup-converge-bio-pulls-in-25m-from-bessemer-and-execs-from-meta-openai-and-wiz/", "sourceType": "press", "accessed": "2026-05-13", "notes": "Independent coverage of Converge Bio's Series A, customer-facing systems, and work across antibody design, protein yield optimization, and biomarker/target discovery.", "source_tier": "C", "publisher_type": "media" }, { "id": "alven-generare-series-a-202604", "title": "Generare raises €20M after generating more novel high quality molecules in 2025 than the rest of the field combined", "publisher": "Alven", "date": "2026-04-02", "url": "https://alven.co/news/generare-raises-20m-after-generating-more-novel-high-quality-molecules-in-2025-than-the-rest-of-the-field-combined", "sourceType": "official", "accessed": "2026-05-13", "notes": "Investor-backed primary release reporting Generare's €20M Series A, Alven and Daphni co-leads, 200 previously uncharacterized small molecules, and plan to scale the dataset ten-fold by 2027.", "source_tier": "B", "publisher_type": "investor" }, { "id": "biologydigital-generare-series-a-202604", "title": "Generare Secures €20M Series A for AI-Driven Drug Discovery", "publisher": "Biology Digital", "date": "2026-04-02", "url": "https://www.biology.digital/news/daily-generare-secures-20m-series-a-for-ai-driven-drug-d-o1dlq-2026-04-02", "sourceType": "press", "accessed": "2026-05-13", "notes": "Independent coverage of Generare's Series A and platform for discovering novel chemistry from microbial genomes.", "source_tier": "C", "publisher_type": "media" }, { "id": "businesswire-alloy-series-e-202604", "title": "Alloy Therapeutics Announces $40M Series E to Scale Tech-Enabled Biotech Infrastructure", "publisher": "Alloy Therapeutics via Business Wire", "date": "2026-04-15", "url": "https://finance.yahoo.com/sectors/healthcare/articles/alloy-therapeutics-announces-40m-series-100000791.html", "sourceType": "official", "accessed": "2026-05-13", "notes": "Press release reporting Alloy's $40M Series E, $1B valuation, 200+ partners, 100+ licensed therapeutic programs, 22 clinical programs, and AI plus wet-lab infrastructure positioning.", "source_tier": "B", "publisher_type": "company_press_release" }, { "id": "citybiz-alloy-series-e-202604", "title": "Alloy Therapeutics Raises $40M Series E at $1B Valuation to Expand AI-Driven Drug Development Platform", "publisher": "citybiz", "date": "2026-04-15", "url": "https://www.citybiz.co/article/832584/alloy-therapeutics-raises-40m-series-e-at-1b-valuation-to-expand-ai-driven-drug-development-platform/", "sourceType": "press", "accessed": "2026-05-13", "notes": "Independent coverage of Alloy's Series E and full-stack biotech infrastructure positioning.", "source_tier": "C", "publisher_type": "media" }, { "id": "ft-project-prometheus-202604", "title": "Jeff Bezos's AI lab nears $38bn valuation in funding deal", "publisher": "Financial Times", "date": "2026-04-21", "url": "https://www.ft.com/content/87ea0ced-bf3c-4822-8dda-437241570ded", "sourceType": "press", "accessed": "2026-05-13", "notes": "Financial Times report on Project Prometheus seeking roughly $10B at a $38B valuation for physical AI focused on engineering and manufacturing.", "source_tier": "C", "publisher_type": "media" }, { "id": "investing-project-prometheus-202604", "title": "Bezos AI lab Project Prometheus raises $10 billion funding round", "publisher": "Investing.com", "date": "2026-04-23", "url": "https://www.investing.com/news/company-news/bezos-ai-lab-project-prometheus-raises-10-billion-funding-round-93CH-4634215", "sourceType": "press", "accessed": "2026-05-13", "notes": "Secondary coverage citing Bloomberg reporting that Project Prometheus closed a $10B round at roughly $38B valuation with JPMorgan and BlackRock among participants.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-scivity-home-202605", "title": "Scivity official site", "publisher": "Scivity Labs", "date": "2026-05", "url": "https://scivity.org/", "sourceType": "official", "accessed": "2026-05-13", "notes": "Primary source for Scivity's public positioning as autonomous research for computational sciences, founded in 2025 in Yerevan, and computational-domain rollout claims.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-scivity-verification-202603", "title": "Verification", "publisher": "Scivity Labs", "date": "2026-03", "url": "https://scivity.org/verification", "sourceType": "official", "accessed": "2026-05-13", "notes": "Primary source for Scivity's public verification framing, including validation outcomes and the claim that verification is core to the product thesis.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-scivity-changelog-202604", "title": "Updates", "publisher": "Scivity Labs", "date": "2026-04", "url": "https://scivity.org/changelog", "sourceType": "official", "accessed": "2026-05-13", "notes": "Primary source for public milestones including site launch, verification work, and the first disclosed end-to-end autonomous research run.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-tetsuwan-home-2026", "title": "Tetsuwan Scientific official site", "publisher": "Tetsuwan Scientific", "date": "2026", "url": "https://tetsuwan.com/", "sourceType": "official", "accessed": "2026-05-13", "notes": "Primary source for Tetsuwan's ResearchOS positioning as the interface between researchers, agents, and automated biology labs, plus its cloud-lab ambition.", "source_tier": "B", "publisher_type": "company" }, { "id": "synbiobeta-tetsuwan-preseed-202411", "title": "First Autonomous Robotic Scientist Hits Labs as Tetsuwan Scientific Secures $2.7M", "publisher": "SynBioBeta", "date": "2024-11-19", "url": "https://www.synbiobeta.com/read/first-autonomous-robotic-scientist-launched-as-tetsuwan-scientific-secures-2-7m", "sourceType": "press", "accessed": "2026-05-13", "notes": "Independent coverage of Tetsuwan's $2.7M pre-seed, investor set, and first reported rare-disease-lab deployment.", "source_tier": "C", "publisher_type": "media" }, { "id": "techcrunch-tetsuwan-preseed-202412", "title": "Tetsuwan Scientific is making robotic AI scientists that can run experiments on their own", "publisher": "TechCrunch", "date": "2024-12-22", "url": "https://techcrunch.com/2024/12/22/tetsuwan-scientific-is-making-robotic-ai-scientists-that-can-run-experiments-on-their-own/", "sourceType": "press", "accessed": "2026-05-13", "notes": "Independent profile of Tetsuwan covering the $2.7M pre-seed, cloud-lab vision, and autonomous wet-lab positioning.", "source_tier": "C", "publisher_type": "media" }, { "id": "crunchbase-tetsuwan-financials-202404", "title": "Tetsuwan Scientific - Financial Details", "publisher": "Crunchbase", "date": "2024-04-18", "url": "https://www.crunchbase.com/organization/tetsuwan-scientific/company_financials", "sourceType": "database", "accessed": "2026-05-13", "notes": "Structured company profile used to confirm Apr 18, 2024 as the announced pre-seed round date and 2048 Ventures as lead investor.", "source_tier": "C", "publisher_type": "company_database" }, { "id": "official-microsoft-discovery-solution-202604", "title": "Microsoft Discovery", "publisher": "Microsoft Azure", "date": "2026-04", "url": "https://azure.microsoft.com/en-us/solutions/discovery", "sourceType": "official", "accessed": "2026-05-13", "notes": "Product page for Microsoft Discovery preview, including the Discovery Engine, enterprise governance claims, and end-to-end R&D lifecycle positioning.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-microsoft-discovery-intro-202505", "title": "Transforming R&D with agentic AI: Introducing Microsoft Discovery", "publisher": "Microsoft Azure Blog", "date": "2025-05-19", "url": "https://azure.microsoft.com/en-us/blog/transforming-rd-with-agentic-ai-introducing-microsoft-discovery/", "sourceType": "official", "accessed": "2026-05-13", "notes": "Launch announcement for Microsoft Discovery at Build 2025, covering the graph-based knowledge engine, hypothesis generation, simulation, and extensibility claims.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-microsoft-discovery-preview-202604", "title": "Microsoft Discovery: Advancing agentic R&D at scale", "publisher": "Microsoft Azure Blog", "date": "2026-04-22", "url": "https://azure.microsoft.com/en-us/blog/microsoft-discovery-advancing-agentic-rd-at-scale/", "sourceType": "official", "accessed": "2026-05-13", "notes": "Expanded preview update for Microsoft Discovery covering customer deployments, broader partner integrations, and physical-lab interoperability claims.", "source_tier": "B", "publisher_type": "company" }, { "id": "techcrunch-microsoft-discovery-202505", "title": "Microsoft wants to tap AI to accelerate scientific discovery", "publisher": "TechCrunch", "date": "2025-05-19", "url": "https://techcrunch.com/2025/05/19/microsoft-wants-to-tap-ai-to-accelerate-scientific-discovery/", "sourceType": "press", "accessed": "2026-05-13", "notes": "Independent coverage of Microsoft's Build 2025 launch of Microsoft Discovery and how it compares to other AI-for-science efforts.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-edison-paperqa3-202602", "title": "Introducing PaperQA3: a frontier multimodal deep research agent for science", "publisher": "Edison Scientific", "date": "2026-02-18", "url": "https://edisonscientific.com/articles/edison-literature-agent", "sourceType": "official", "accessed": "2026-05-13", "notes": "Primary source for PaperQA3, Edison Literature availability, multimodal figure-and-table reasoning, and benchmark claims against LABBench2 and HLE subsets.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-edison-nvidia-202603", "title": "Accelerating Science at Scale", "publisher": "Edison Scientific", "date": "2026-03-16", "url": "https://edisonscientific.com/articles/accelerating-science-at-scale", "sourceType": "official", "accessed": "2026-05-13", "notes": "Primary source for Edison's NVIDIA partnership narrative, BixBench-Hypothesis release, and updated claims around Kosmos scale and reproducibility.", "source_tier": "B", "publisher_type": "company" }, { "id": "nvidia-edison-nemotron-case-study-202603", "title": "Scaling Multimodal Scientific Literature Understanding with NVIDIA Nemotron Parse", "publisher": "NVIDIA Developer", "date": "2026-03", "url": "https://developer.nvidia.com/case-studies/scientific-literature-ai-nvidia-nemotron", "sourceType": "case_study", "accessed": "2026-05-13", "notes": "Partner case study describing Edison's Literature agent, Kosmos deployment scale, and Nemotron-backed multimodal parsing improvements.", "source_tier": "B", "publisher_type": "technology_partner" }, { "id": "official-anthropic-claude-life-sciences-202510", "title": "Claude for Life Sciences", "publisher": "Anthropic", "date": "2025-10-20", "url": "https://www.anthropic.com/news/claude-for-life-sciences", "sourceType": "official", "accessed": "2026-05-14", "notes": "Anthropic launch announcement for Claude for Life Sciences, including Benchling, PubMed, 10x Genomics, ClinicalTrials.gov, and Synapse connectors, Agent Skills, and pharma partners Sanofi, Novo Nordisk, AbbVie, AstraZeneca.", "source_tier": "B", "publisher_type": "company" }, { "id": "official-anthropic-allen-hhmi-202510", "title": "Anthropic partners with Allen Institute and Howard Hughes Medical Institute", "publisher": "Anthropic", "date": "2025-10-20", "url": "https://www.anthropic.com/news/anthropic-partners-with-allen-institute-and-howard-hughes-medical-institute", "sourceType": "official", "accessed": "2026-05-14", "notes": "Anthropic announcement of founding life-sciences partnerships with the Allen Institute and HHMI extending Claude to frontier scientific research workflows.", "source_tier": "B", "publisher_type": "company" }, { "id": "biospace-anthropic-coefficient-202604", "title": "AI Giant Anthropic Leans Into Life Sciences With $400M Coefficient Bio Catch", "publisher": "BioSpace", "date": "2026-04-04", "url": "https://www.biospace.com/business/ai-giant-anthropic-leans-into-life-sciences-with-400m-coefficient-bio-catch", "sourceType": "press", "accessed": "2026-05-14", "notes": "Independent BioSpace coverage confirming the Anthropic / Coefficient Bio acquisition, all-stock structure, team composition, and life-sciences strategic intent.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-ai2-asta-launch-202508", "title": "Asta: Accelerating science through trustworthy agentic AI", "publisher": "Ai2", "date": "2025-08-26", "url": "https://allenai.org/blog/asta", "sourceType": "official", "accessed": "2026-05-14", "notes": "Ai2 launch post for Asta covering the open agentic ecosystem for science: Asta agents, AstaBench, and Asta Resources developer toolkit with the Scientific Corpus Tool.", "source_tier": "B", "publisher_type": "nonprofit_research" }, { "id": "official-ai2-asta-platform", "title": "Ai2 Asta", "publisher": "Ai2", "date": "2026", "url": "https://asta.allen.ai/", "sourceType": "official", "accessed": "2026-05-14", "notes": "Primary Asta platform page describing scholarly research assistants, agent stack, and links to AstaBench and Ai2 research products.", "source_tier": "B", "publisher_type": "nonprofit_research" }, { "id": "businesswire-ai2-asta-202508", "title": "Ai2 Launches Asta: a New Standard for Trustworthy AI Agents in Science", "publisher": "Business Wire", "date": "2025-08-26", "url": "https://www.businesswire.com/news/home/20250826827940/en/Ai2-Launches-Asta-a-New-Standard-for-Trustworthy-AI-Agents-in-Science", "sourceType": "press", "accessed": "2026-05-14", "notes": "Independent press-release distribution coverage of the Ai2 Asta launch, scope of AstaBench, and Ai2 nonprofit positioning.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-aws-bio-discovery-blog-202604", "title": "Applying multimodal biological foundation models across therapeutics and patient care", "publisher": "AWS", "date": "2026-04-28", "url": "https://aws.amazon.com/blogs/machine-learning/applying-multimodal-biological-foundation-models-across-therapeutics-and-patient-care/", "sourceType": "official", "accessed": "2026-05-14", "notes": "AWS launch blog for Amazon Bio Discovery describing 40+ biology foundation models, lab-in-the-loop orchestration with Bedrock agents, CRO and cloud-lab partner network, and Memorial Sloan Kettering early antibody project.", "source_tier": "B", "publisher_type": "company" }, { "id": "genengnews-aws-bio-discovery-202604", "title": "AWS Launches Amazon Bio Discovery Agentic AI to Accelerate Drug Development", "publisher": "GEN — Genetic Engineering & Biotechnology News", "date": "2026-04-29", "url": "https://www.genengnews.com/topics/artificial-intelligence/aws-launches-amazon-bio-discovery-agentic-ai-to-accelerate-drug-development/", "sourceType": "press", "accessed": "2026-05-14", "notes": "Industry coverage confirming Amazon Bio Discovery agentic AI launch, partner ecosystem (Ginkgo, Twist Bioscience, A-Alpha Bio), and BioFM catalog scope.", "source_tier": "C", "publisher_type": "media" }, { "id": "aipedia-aws-bio-discovery-202604", "title": "AWS launches Amazon Bio Discovery for AI-assisted drug research", "publisher": "AIpedia", "date": "2026-04-28", "url": "https://www.aipedia.wiki/news/2026-04-28-amazon-bio-discovery-drug-research/", "sourceType": "press", "accessed": "2026-05-14", "notes": "Additional independent coverage of the Amazon Bio Discovery launch and lab-in-the-loop architecture summary.", "source_tier": "C", "publisher_type": "media" }, { "id": "official-profluent", "title": "Profluent official site", "publisher": "Profluent", "date": "2026", "url": "https://www.profluent.bio/", "sourceType": "official", "accessed": "2026-05-14", "notes": "Primary Profluent website with product, research, and team identity, including OpenCRISPR and protein foundation model positioning.", "source_tier": "B", "publisher_type": "company" }, { "id": "businesswire-profluent-series-b-202511", "title": "Profluent Raises $106M to Scale Frontier AI Models for Programmable Biology", "publisher": "Business Wire", "date": "2025-11-19", "url": "https://www.businesswire.com/news/home/20251119356889/en/Profluent-Raises-$106M-to-Scale-Frontier-AI-Models-for-Programmable-Biology", "sourceType": "official", "accessed": "2026-05-14", "notes": "Primary Series B announcement: $106M co-led by Altimeter Capital and Bezos Expeditions, $150M total funding, OpenCRISPR-1 adoption, scaling-laws NeurIPS 2025 spotlight.", "source_tier": "B", "publisher_type": "company_press_release" }, { "id": "biospace-profluent-lilly-202604", "title": "Lilly, AI biotech Profluent ink $2.25B pact in search of genetic medicine 'holy grail'", "publisher": "BioSpace", "date": "2026-05-04", "url": "https://www.biospace.com/deals/lilly-ai-biotech-profluent-ink-2-25b-pact-in-search-of-genetic-medicine-holy-grail", "sourceType": "press", "accessed": "2026-05-14", "notes": "Independent BioSpace coverage of the Profluent / Eli Lilly strategic collaboration for AI-designed site-specific recombinases, with up to $2.25B in milestones plus tiered royalties.", "source_tier": "C", "publisher_type": "media" }, { "id": "airstreet-profluent-lilly-202604", "title": "Profluent and Lilly partner on $2.25B+ gene editing deal", "publisher": "Air Street Press", "date": "2026-05-04", "url": "https://press.airstreet.com/p/profluent-lilly", "sourceType": "press", "accessed": "2026-05-14", "notes": "Investor-side commentary on the Profluent / Lilly recombinase deal, milestone structure, and strategic context within the genetic medicine market.", "source_tier": "C", "publisher_type": "investor" }, { "id": "anthropic-claude-mythos-202604", "title": "Claude Mythos Preview", "publisher": "Anthropic", "date": "2026-04-07", "url": "https://red.anthropic.com/2026/mythos-preview/", "sourceType": "official", "accessed": "2026-05-29", "notes": "Anthropic preview of Claude Mythos, positioned a capability tier above Opus 4.7 and initially limited to the invitation-only Project Glasswing program; basis for treating Mythos as the current frontier-capability ceiling. Reported scores: 93.9% SWE-bench Verified, 77.8% SWE-bench Pro, 82% Terminal-Bench 2.0, and 97.6% USAMO 2026, each a double-digit lead over Opus 4.6 and GPT-5.4. Anthropic says Mythos-class models will roll out to all customers shortly.", "source_tier": "B", "publisher_type": "company" }, { "id": "axios-anthropic-opus48-202605", "title": "Anthropic releases new model, Opus 4.8", "publisher": "Axios", "date": "2026-05-28", "url": "https://www.axios.com/2026/05/28/anthropic-opus-release-mythos", "sourceType": "media", "accessed": "2026-05-29", "notes": "News coverage of the Claude Opus 4.8 general-availability release and the staged expansion of Mythos-class access.", "source_tier": "C", "publisher_type": "media" }, { "id": "artificial-analysis-intelligence-index", "title": "Artificial Analysis Intelligence Index", "publisher": "Artificial Analysis", "date": "2026-05", "url": "https://artificialanalysis.ai/evaluations/artificial-analysis-intelligence-index", "sourceType": "benchmark", "accessed": "2026-05-29", "notes": "Composite intelligence index (v4.0) aggregating ten evaluations including GPQA Diamond, Humanity's Last Exam, SciCode, and Terminal-Bench Hard; used for current frontier-model ordering (Claude Opus 4.8, GPT-5.5, Gemini 3.1 Pro).", "source_tier": "B", "publisher_type": "benchmark_provider" }, { "id": "epoch-eci-gpt55pro-202604", "title": "GPT-5.5 Pro achieves a new high score on the ECI", "publisher": "Epoch AI", "date": "2026-04-28", "url": "https://epochai.substack.com/p/gpt-55-pro-achieves-a-new-high-score", "sourceType": "benchmark", "accessed": "2026-05-29", "notes": "Epoch AI report that GPT-5.5 Pro reached a new high of 159 on the Epoch Capabilities Index (90% CI 156-162), on a scale calibrated so GPT-5 = 150 and Claude 3.5 Sonnet = 130.", "source_tier": "B", "publisher_type": "benchmark_provider" }, { "id": "arcprize-arc-agi2-leaderboard-202605", "title": "ARC-AGI Leaderboard", "publisher": "ARC Prize Foundation", "date": "2026-05", "url": "https://arcprize.org/leaderboard", "sourceType": "benchmark", "accessed": "2026-05-29", "notes": "Official ARC Prize leaderboard for ARC-AGI-1 and ARC-AGI-2, which measure abstraction and generalization on novel tasks designed to resist memorization.", "source_tier": "B", "publisher_type": "benchmark_provider" }, { "id": "llm-stats-arc-agi2-202605", "title": "ARC-AGI v2 Benchmark Leaderboard", "publisher": "LLM-Stats", "date": "2026-05", "url": "https://llm-stats.com/benchmarks/arc-agi-v2", "sourceType": "benchmark", "accessed": "2026-05-29", "notes": "Aggregated ARC-AGI-2 model scores: GPT-5.5 0.850, Gemini 3.1 Pro 0.771, GPT-5.4 0.733, Gemini 3.5 Flash 0.721, Claude Opus 4.6 0.688.", "source_tier": "C", "publisher_type": "aggregator" }, { "id": "codeant-swebench-leaderboard-202602", "title": "SWE-bench Leaderboard 2026: All Model Scores, Rankings & What They Actually Mean", "publisher": "CodeAnt AI", "date": "2026-02", "url": "https://www.codeant.ai/blogs/swe-bench-scores", "sourceType": "media", "accessed": "2026-05-29", "notes": "Secondary aggregation of SWE-bench Verified and SWE-bench Pro scores, including the contamination gap between Verified (~81%) and Pro (~46%) for the same models.", "source_tier": "C", "publisher_type": "media" }, { "id": "morphllm-swebench-pro-202603", "title": "SWE-Bench Pro Leaderboard (2026): Why 46% Beats 81%", "publisher": "Morph", "date": "2026-03", "url": "https://www.morphllm.com/swe-bench-pro", "sourceType": "benchmark", "accessed": "2026-05-29", "notes": "Analysis of the SEAL SWE-bench Pro leaderboard (1,865 multi-language, contamination-resistant tasks): Claude Opus 4.5 45.9%, Claude Sonnet 4.5 43.6%, Gemini 3 Pro 43.3%, versus ~81% on the easier Verified split.", "source_tier": "C", "publisher_type": "aggregator" }, { "id": "stanford-ai-index-2026", "title": "The 2026 AI Index Report", "publisher": "Stanford HAI", "date": "2026-04", "url": "https://hai.stanford.edu/ai-index/2026-ai-index-report", "sourceType": "report", "accessed": "2026-05-29", "notes": "Stanford HAI annual AI Index: frontier models now match or exceed expert humans on PhD-level science (GPQA ~93% vs ~81% validator baseline) and agent task success climbed sharply (Terminal-Bench 20% to 77.3%), while remaining unreliable in several scientific domains.", "source_tier": "B", "publisher_type": "academic_or_lab" }, { "id": "metr-mythos-time-horizon-202605", "title": "METR Task-Completion Time Horizons — Claude Mythos Preview", "publisher": "METR", "date": "2026-05", "url": "https://metr.org/time-horizons/", "sourceType": "research", "accessed": "2026-06-01", "notes": "METR added Claude Mythos Preview (early) on May 8, 2026; its 50% time horizon lands above all prior models, past the ~16-hour ceiling where the current task suite gives reliable measurements. Opus 4.6 and GPT-5.2 cluster around 5-6 hours; post-2023 doubling time is ~4-5 months (TH1.1 ~130 days).", "source_tier": "B", "publisher_type": "nonprofit_research" }, { "id": "google-gemini35-flash-202605", "title": "Gemini 3.5: frontier intelligence with action", "publisher": "Google", "date": "2026-05", "url": "https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-5/", "sourceType": "official", "accessed": "2026-06-01", "notes": "Gemini 3.5 Flash (released May 19, 2026) reaches frontier-level intelligence at ~4x output speed: 76.2% Terminal-Bench 2.1 (vs 70.3% for Gemini 3.1 Pro), 83.6% MCP Atlas, 57.9% Finance Agent v2, and 1656 Elo on GDPval-AA, priced at $1.50/$9 per 1M tokens. Gemini 3.5 Pro expected mid-2026.", "source_tier": "B", "publisher_type": "company" }, { "id": "arxiv-firebench-202602", "title": "FIRE-Bench: Evaluating Agents on the Rediscovery of Scientific Insights", "publisher": "arXiv", "date": "2026-02", "url": "https://arxiv.org/pdf/2602.02905", "sourceType": "research", "accessed": "2026-06-01", "notes": "FIRE-Bench gives an agent only a high-level research question and requires it to autonomously explore ideas, design experiments, implement and execute code, and derive conclusions. Even the strongest agents achieve limited rediscovery success, showing full-cycle scientific research remains hard.", "source_tier": "B", "publisher_type": "preprint_server" }, { "id": "arxiv-autoresearchbench-202604", "title": "AutoResearchBench: Benchmarking AI Agents on Complex Scientific Literature Discovery", "publisher": "arXiv", "date": "2026-04", "url": "https://arxiv.org/html/2604.25256v1", "sourceType": "research", "accessed": "2026-06-01", "notes": "AutoResearchBench tests whether an agent can search a large, up-to-date scientific corpus, read full papers in depth, verify fine-grained technical conditions, and decide when its search is complete.", "source_tier": "B", "publisher_type": "preprint_server" }, { "id": "metr-measuring-long-tasks-202503", "title": "Measuring AI Ability to Complete Long Tasks", "publisher": "METR", "date": "2025-03", "url": "https://metr.org/blog/2025-03-19-measuring-ai-ability-to-complete-long-tasks/", "sourceType": "research", "accessed": "2026-06-01", "notes": "Original METR study introducing the task-completion time horizon: the human-expert task length a model finishes at 50% reliability. Reports a ~7-month doubling over 6 years (Claude 3.7 Sonnet = 59 min) and is the foundation for the exponential-trend framing.", "source_tier": "B", "publisher_type": "nonprofit_research" }, { "id": "aidigest-time-horizons", "title": "A new Moore's Law for AI agents", "publisher": "AI Digest", "date": "2026-05", "url": "https://theaidigest.org/time-horizons", "sourceType": "industry-analysis", "accessed": "2026-06-01", "notes": "Interactive tracker of METR-style time horizons, framing the doubling trend as a Moore's Law for AI agents and visualizing the extrapolation toward day-, week-, and month-long autonomous tasks.", "source_tier": "C", "publisher_type": "industry_analysis" }, { "id": "arxiv-rebench-202411", "title": "RE-Bench: Evaluating frontier AI R&D capabilities of language model agents against human experts", "publisher": "arXiv", "date": "2024-11", "url": "https://arxiv.org/pdf/2411.15114", "sourceType": "research", "accessed": "2026-06-01", "notes": "Seven open-ended ML research-engineering environments with data from 71 eight-hour attempts by 61 human experts. The best AI agents score ~4x higher than human experts when both are limited to a 2-hour budget per environment, a direct measurement of AI contributing to AI R&D.", "source_tier": "B", "publisher_type": "preprint_server" }, { "id": "arxiv-compute-horizons-202511", "title": "Forecasting AI Time Horizon Under Compute Slowdowns", "publisher": "arXiv", "date": "2025-11", "url": "https://arxiv.org/pdf/2511.19492", "sourceType": "research", "accessed": "2026-06-01", "notes": "Follow-up analysis showing that accounting for plausible compute-scaling slowdowns pushes the 1-month (80% reliability) task horizon roughly 7 years later than the naive trend extrapolation. Used as the uncertainty caveat on the extrapolation.", "source_tier": "B", "publisher_type": "preprint_server" }, { "id": "axios-anthropic-rsi-202605", "title": "Behind the Curtain: Intelligence explosion", "publisher": "Axios", "date": "2026-05-07", "url": "https://www.axios.com/2026/05/07/anthropic-jack-clark-ai-intelligence-explosion", "sourceType": "press", "accessed": "2026-06-01", "notes": "Reports Anthropic stating in writing that it sees AI contributing to speeding up the R&D of AI itself (early recursive self-improvement), that a majority of its code is now written by Claude Code, and that it will publish more on how AI tools have accelerated its own work.", "source_tier": "C", "publisher_type": "media" }, { "id": "techcrunch-rsi-202605", "title": "RSI is the new AGI — and it's just as hard to pin down", "publisher": "TechCrunch", "date": "2026-05-28", "url": "https://techcrunch.com/2026/05/28/rsi-is-the-new-agi-and-its-just-as-hard-to-pin-down/", "sourceType": "press", "accessed": "2026-06-01", "notes": "Surveys recursive self-improvement claims, including Jack Clark's estimate of a greater-than-even chance by end of 2028 that a system could autonomously make a better version of itself. Used to label the forecast as a forecast, not a measurement.", "source_tier": "C", "publisher_type": "media" }, { "id": "anthropic-institute-agenda-202603", "title": "Focus areas for The Anthropic Institute", "publisher": "Anthropic", "date": "2026-03", "url": "https://www.anthropic.com/research/anthropic-institute-agenda", "sourceType": "official", "accessed": "2026-06-01", "notes": "Anthropic's research institute (March 2026) studying how powerful AI reshapes the economy, security, and society, including the implications of potential recursive self-improvement of AI systems.", "source_tier": "B", "publisher_type": "company" }, { "id": "arxiv-gpqa-202311", "title": "GPQA: A Graduate-Level Google-Proof Q&A Benchmark", "publisher": "arXiv", "date": "2023-11", "url": "https://arxiv.org/pdf/2311.12022", "sourceType": "research", "accessed": "2026-06-01", "notes": "Source for the human expert baseline on GPQA: domain PhD experts score roughly 65-74% (and ~81% on the validated Diamond subset), the reference line frontier models have since surpassed (~94%).", "source_tier": "B", "publisher_type": "preprint_server" }, { "id": "arxiv-osworld-202404", "title": "OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments", "publisher": "arXiv", "date": "2024-04", "url": "https://arxiv.org/pdf/2404.07972", "sourceType": "research", "accessed": "2026-06-01", "notes": "Real-computer-use agent benchmark; human performance is around 72%, the reference point against which 2026 frontier agents reach approximate parity (~75%).", "source_tier": "B", "publisher_type": "preprint_server" } ] }