For centuries, the scientific method has been the greatest engine of human progress. At Google, our mission is deeply rooted in building tools that accelerate it. We believe that the novel era of discovery will not come thanks to narrow, specialized models, but general factors that will empower researchers from all fields of science.
That’s why we present Gemini for sciencea collection of scientific tools and experiments designed to escalate the scale and precision of scientific research.
A force multiplier for human ingenuity
Today, science faces a paradox: our collective knowledge is growing so rapidly that it is becoming increasingly arduous for individual scientists to see the massive picture. Scientific breakthroughs often involve making innovative connections between data, but the time it takes to do this manually can take weeks or even months. Artificial intelligence can lend a hand eliminate this bottleneck and serve as a force multiplier in scientific work by performing intricate tasks. This allows researchers to focus on identifying and solving the most vital scientific problems and directions that will drive progress.
The Gemini for Science experimental tools in Google Labs include three main prototypes designed to handle such tasks.
- Hypothesis generation,constructed using Co-scientist: : Ideations are the heart of science, but no one human can synthesize the millions of papers published every year. Hypothesis Generation fills this gap by simulating the scientific method: it works with researchers to define a research challenge, then uses a multi-agent “tournament of ideas” to generate, debate, and evaluate hypotheses. To ensure absolute rigor, claims are thoroughly verified and supported by clickable quotes.
- Computational discovery,built using AlphaEvolution AND ERA (Empirical Research Assistance): : Scientific progress is often restricted by the number of hypotheses we can realistically test through computational experiments. Computational Discovery, an agent-based research engine, is a prototype that solves this problem by generating and evaluating thousands of code variations in parallel. This enables scientists to test novel modeling approaches – for intricate fields such as solar energy forecasting or epidemiology – that would take months to navigate manually.
- Literature Insights, built from GoogleLM Notebook: : Understanding the scientific literature is a fundamental part of all research journeys. Literature Insights searches the scientific literature and organizes the results into tables with custom, searchable attributes for side-by-side analysis. Researchers can apply chat to discover nuances contained in their selected corpus and create high-quality artifacts such as reports, slide decks, infographics, and audio and video overviews. With the power of NotebookLM, the literature analysis feature helps you synthesize findings from different articles, identify research gaps, and discover areas of opportunity.
Starting today, we will gradually open access to these experiments. Visit labs.google/science to register your interest.
In addition to individual experiments, we also provide advanced artificial intelligence capabilities to corporate organizations Google Cloud. Our enterprise-class solutions for scientific and industrial research are already being used by many partners in private preview to make a real-world impact. Companies like BASF they apply AlphaEvolution optimizing their supply chains, and Klarna uses them to improve its machine learning models. In parallel, organizations such as Daiichi Sankyo, Bayer Crop Science and US National Labs (part of the US Department of Energy’s Genesis Mission) are using the Co-Scientist program to accelerate their research and address fundamental science challenges. These enterprise-grade tools are showing significant value in the current preview phase. We’re excited about the breakthroughs our partners are unlocking and look forward to expanding access to more organizations in the coming months.
Several validation papers based on these and other tools have already been published. The ERA AND Co-scientist The research papers were published today in the journal Nature.
Scientific dashboard on your desktop
As part of Gemini for Science, we are also launching Science Skills, a specialized suite integrating insights from over 30 major life science databases and tools, including UniProt, AlphaFold DatabaseAlphaGenome API i InterPro. Using these skills on agent-based platforms like Google Antigravity enables researchers to perform intricate and often manual processes like structural bioinformatics and genomic analyzes in minutes, not hours.
Our research teams using Science Skills have already seen this acceleration in practice. In the early stages of testing, our team used Science Skills to perform intricate analysis that typically takes hours and minutes. This has led to novel insights into the potential mechanisms of a uncommon genetic disease caused by mutations in the AK2 gene.