Gemini for Science: AI tools and experiments to accelerate research

Gemini for Science is a collection of AI tools and experiments designed to expand the scale and precision of scientific exploration, aiming to equip researchers across disciplines with general agents that handle complex tasks and free humans to pursue the most impactful problems.

Three prototype tools are available as experiments. Hypothesis Generation, built with Co-Scientist, runs a multi-agent “idea tournament” to generate, debate and evaluate hypotheses with deep verification and clickable citations. Computational Discovery, built with AlphaEvolve and ERA, generates and scores thousands of code variations in parallel so researchers can test many computational experiments at once.

Literature Insights, built with Google NotebookLM, searches and structures scientific literature into searchable tables and supports chat-driven synthesis and high-fidelity artifacts like reports and slide decks.

gemini, ai tools, hypothesis generation, co-scientist, computational discovery, alphaevolve, era, literature insights, notebooklm, multi-agent