Biohub expands AI cell research into a $1.8 billion effort

Biohub, the science nonprofit founded by Mark Zuckerberg and Priscilla Chan, announced on October 7 that it is expanding its Virtual Biology Initiative into a $1.8 billion effort. Federal agencies, Google DeepMind, Meta and Isomorphic Labs are joining the effort to gather data for AI that can simulate cells.

The goal is a “universal virtual cell,” AI software that would predict how cells respond to a drug or another change before a lab test. As The Rundown reported, Biohub is backing that ambition with funding, scientific data and research talent.

What the $1.8 billion covers

The total combines funding, data, computing and measurement technology, including resources built with past public spending. Biohub’s original $500 million initiative, announced April 29, runs over five years. It allocated $400 million to its own technology and data generation and $100 million to external research.

The National Institutes of Health is contributing datasets, repositories and knowledge bases developed through more than $500 million in previous federal investment. Biohub will help standardize those resources for AI training.

The Department of Energy has committed more than $500 million over five years for biological measurements, imaging, modeling, data collection and computation. It plans to bring supercomputers, experimental facilities and automated labs to the work.

Google DeepMind, Isomorphic Labs and Meta have collectively committed $300 million. Biohub’s announcement does not disclose the individual contributions.

Why it matters

On October 6, OpenAI reported mathematical results from an internal model. Biohub’s expansion puts biology among AI’s next big scientific targets. Its mission to “cure or prevent all disease” makes this an AI megaproject with a clear public purpose, and one that is easier to root for than many others.

Biohub also announced in November 2025 that EvolutionaryScale’s team would join, with Alex Rives becoming head of science. The work brings experimental biology, data collection and AI research together.

For scientists studying disease and teams developing drugs, the practical benefit could be better choices about which experiments to run. A reliable cell model could help them decide which drugs or other changes to test first.

Biohub says it needs much more biological data. That makes the planned imaging and automated lab work central to the AI effort. Better observations of cells could supply the material for training models and testing their predictions.

The benefits could also extend beyond Biohub’s own labs. Funding for external research, reusable datasets and shared standards could give other scientists resources for their own models and experiments. That could improve the tools available for disease research before a universal cell simulator exists.

Rives told Axios that commercial partners receive a year of exclusivity over data they develop before public release. That gives those companies an earlier chance to build on the data, while outside researchers wait for access.

Reproducible cell measurements and model predictions that hold up in lab experiments would show meaningful progress. Researchers would still need to establish how those predictions can translate into medical benefits.