Gemini for Science: AI experiments and tools for a new era of discovery
DeepMind Blog · 2026-05-17
Google DeepMind launches Gemini for Science, a suite of AI research tools—including Co-Scientist for hypothesis generation, AlphaEvolve for computational discovery, and NotebookLM-powered literature analysis—alongside a Science Skills bundle integrating 30+ life science databases, with enterprise partners and research institutions already in private preview.
Appears in
Extraction
Topics: gemini-for-scienceco-scientistalphaevolveai-for-sciencescientific-discovery
Claims
- Gemini for Science comprises three experimental tools: Hypothesis Generation (Co-Scientist), Computational Discovery (AlphaEvolve and ERA), and Literature Insights (NotebookLM), accessible via labs.google/science.
- Science Skills integrates over 30 major life science databases including UniProt, AlphaFold Database, AlphaGenome API, and InterPro for use in agentic platforms like Google Antigravity.
- Google is collaborating with over 100 institutions, including Stanford, Imperial College London, and The Crick Institute, to validate the new systems.
- Enterprise partners including BASF, Klarna, Daiichi Sankyo, and Bayer Crop Science are using these tools in private preview.
- ERA and Co-Scientist research papers are being published in Nature alongside this announcement.
- Google is piloting AI-assisted peer review tools (PAT and ScholarPeer) with conferences including ICML, STOC, and NeurIPS.
Key quotes
We believe that a new era of discovery won't come from narrow, specialized models, but general agents that empower researchers across every scientific field.
Scientific breakthroughs often rely upon making creative connections between data, but the time required to do this manually can take weeks or even months.
AlphaFold has helped over 3 million researchers tackle malaria vaccines and plastic-eating enzymes.