We are inspired by the ability of AI to help tackle the grand challenges in science. At Google, we have a unique opportunity to accelerate scientific progress by using AI to help transform healthcare, advance towards a more sustainable future, and drive breakthroughs in natural sciences. Our research scientists and engineers collaborate with research institutions around the globe and share our progress in scientific publications and open source releases. Together, our aim is to enable scientific innovation for a better world.
Powering a new era of discovery
Gemini for Science is a new collection of science tools and experiments to expand the scale and precision of scientific exploration. Explore a new era of discovery with the Science plugin in Google Antigravity and experimental tools on Google Labs designed to help accelerate core steps of the scientific method.
A multi-agent AI partner to accelerate research
We believe AI can help dramatically accelerate the pace of breakthroughs by serving as a dedicated partner in the generation and refinement of breakthrough scientific hypotheses. Co-Scientist is a new multi-agent system built with Gemini that helps researchers accelerate scientific breakthroughs.
Gary Peltz, a geneticist at Stanford University, is using Co-Scientist to accelerate the search for liver fibrosis treatments. By analyzing biomedical literature, the AI system highlighted overlooked drug-repurposing candidates, including one that successfully blocked 91% of a scarring-linked response in lab tests. The results point toward new gene-regulating approaches to treat chronic liver disease.
Co-Scientist helped unite Ritu Raman and Ryan Flynn’s labs around the degenerative disease, ALS. The system helped Ritu quickly digest complex literature, propose testable ideas, and spot where complementary expertise could strengthen the best leads, sparking collaboration with Ryan on potential RNA-based approaches to ALS.
Biologists Omar Abudayyeh and Jonathan Gootenberg are using Co‑Scientist to speed up research on reversing cellular aging. The system synthesises decades of literature to propose novel genetic leads that in lab tests have been shown to rejuvenate cells. It also slashes the time needed to analyse huge screening datasets, from months to days.
Filippo Menolascina, a bioengineer at the University of Edinburgh, is using Co-Scientist to turn biomedical literature overload into high-quality hypotheses for metabolic liver disease. The system highlighted promising disease mechanisms and drug combinations, and helped explain why an existing drug benefits only some patients — an idea later supported by Menolascina’s lab tests.
At Calico Life Sciences, Matt Onsum and Katherine Labbé are using Co-Scientist to tackle one of medicine’s hardest problems: the biology of aging. The AI system has impressed Calico’s experts with its scientific discernment, including by generating an exciting novel hypothesis about the integrated stress response that was later confirmed in the lab.
Clare Bryant, an immunologist at the University of Cambridge, is using Co-Scientist to help her identify the proteins that cause severe disease when pathogens like flu and Covid-19 leap from animals to humans. Iterating with the AI tool, she rapidly narrowed the hunt to specific amino acids her lab will test — potentially cutting years of experimental work down to months.
Mapping the brain to solve the mysteries of the mind
How does the group of cells that form a brain work together to bring thoughts, movements, and feelings to life? To help answer this question, Google researchers are using AI to create some of the most detailed maps of brains ever made — enabling new insights about how neurons network to share and process information, and revealing never-before-seen discoveries.
Marking ten years of connectomics research at Google, we published in Science the first large-scale reconstruction of a small piece of the human brain at the synaptic level. We presented several new neuron structures discovered in the data, and we opened the dataset and software tools to the scientific community for continued discovery.
AlphaGenome Atlas is the most comprehensive map of single-letter variant changes in the human genome. Atlas simplified variant prioritisation with the AlphaGenome Variant Impact (AVI) score, allowing researchers to quickly prioritize between different variants, while also interpreting molecular effects across all genetic code, both those that produce proteins and control gene activity. It’s available to disease researchers around the world.
Google AI tools are helping scientists decode the genomics of life on Earth, unlocking insights that protect biodiversity and drive medical and environmental breakthroughs. Discover how building this comprehensive catalog of life can redefine our fundamental understanding of the natural world.
Mapping the genes in human DNA is critical for improving the treatment of disease, but for years, the only complete human genome was based on DNA samples scientists collected from a small handful of volunteers in the United States. As part of an international effort to improve upon this single linear genome, our technologies are helping create a pangenome — a record of the genetic diversity of ancestries from around the world.
Uncovering the root causes of disease is one of the greatest challenges in human genetics. With millions of possible mutations and limited experimental data, it’s largely still a mystery which ones could give rise to disease. This knowledge is crucial to faster diagnosis and developing life-saving treatments.
AlphaFold 3 predicts the structure and interactions of all life’s molecules
AlphaFold, an AI technology developed by Google’s DeepMind, is accelerating the pace of scientific innovation by helping researchers understand the intricate 3D shapes of proteins. It heralds a new era in digital biology that is helping to unlock breakthroughs across medicine, agriculture and sustainability.
Dr. Daudi Jjingo and his team at Makerere University are using AlphaFold and AlphaGenome to accelerate research into regional health challenges, including malaria, sickle cell disease, and cancer
Guided by AlphaFold, scientists have determined the first full-length structure of a protein called Pfs48/45, which is a promising candidate for a malaria vaccine. This could help to design new vaccines that induce the production of transmission-blocking antibodies.
Antibiotic resistance causes 2.8M infections in the US alone each year. A team using AlphaFold identified a bacterial protein structure in half an hour that had been elusive for 10 years.
Using AlphaFold and electron microscopy, we are getting closer to understanding the molecular mechanisms related to insect immunity. The results are specifically relevant to honeybee health, which is a topic of global concern due to rapidly declining pollinator numbers.
A team discovering and engineering enhanced enzymes that can eventually be applied at scale to break down some of the most polluting single-use plastics used AF to screen 100 candidate enzymes — too many to generate 3D structures for — in days, giving them a new library of templates to engineer faster, more stable and cheaper enzymes for plastic recycling.
Listening to what birds tell us about biodiversity
Researchers at Google are using AI to build a map of plant, animal, fungi and other species across the world. These advanced Species Distribution Models (SDMs) help scientists, policymakers, and conservationists identify critical habitats, supporting protection efforts where they matter most.
Rich ecosystem or empty forest? Sometimes it’s hard to tell, but getting an answer is critical for scientists working to understand the effects of climate change or the impact of conservation measures. Our researchers have created new tools to measure and understand biodiversity through birdsong.
AI tool GNoME finds 2.2 million new crystals
Modern technologies from computer chips and batteries to solar panels rely on inorganic crystals, which must be stable, otherwise they can decompose. And behind each new stable crystal can be months of painstaking experimentation.
Better, faster predictions of Earth’s weather and climate
Weather and climate patterns are the result of the complicated interactions of many factors — ocean and wind currents, clouds, rain, sun, and topography to name just a few — so it takes some of the world’s largest computers many hours to make the weather predictions we all rely on. Google Researchers are using ML to achieve even better forecasts in just a fraction of the time.
In partnership with the National Hurricane Center, WeatherNext provided critical data to help predict Hurricane Melissa’s rapid Category 5 intensification and Jamaican landfall three days early and with greater accuracy than other models.
Climate change is real, and its effects are becoming increasingly evident. Traditional climate models struggle to produce clear, accurate pictures of how our climate will continue to change due to limitations in how the models represent complex phenomena. NeuralGCM is a groundbreaking AI-powered approach that could someday offer a faster, more efficient, and more accurate way to predict climate change.
The ionosphere, a region in Earth’s upper atmosphere, is a swirling sea of charged particles. In a new study published in Nature, we report the use of aggregated sensor measurements from millions of Android phones to map the ionosphere at a level of accuracy that matches or far exceeds that of conventional monitoring infrastructure.
Earthquakes are a constant threat to communities around the globe. Using aggregated measurements from a global network of Android smartphones, we developed a system that detects earthquakes, delivers early warnings to users, and builds user trust with each successful alert.
AlphaGeometry
AI systems often struggle with complex problems in geometry and mathematics due to a lack of reasoning skills and training data. AlphaGeometry’s system combines the predictive power of a neural language model with a rule-bound deduction engine, which work in tandem to find solutions.
