The U.S. government and technology companies including Meta Platforms and Google are partnering with nonprofit Biohub to develop open datasets designed to train artificial intelligence models for biological research, bringing total investment in the initiative to $1.8 billion.
The project aims to create large-scale biological datasets that can help AI systems better understand how cells function and accelerate research in areas such as drug discovery and medical science.
Major Technology and Government Funding
Meta, Google DeepMind and drug discovery startup Isomorphic Labs are jointly investing $300 million into the initiative.
The U.S. Department of Energy will provide more than $500 million over five years to support laboratory measurements, scientific modeling and computational research. The National Institutes of Health will coordinate datasets and repositories created through more than $500 million in previous federal funding, which Biohub will standardize for AI training.
The new commitments build on the $500 million investment Biohub announced in April. The nonprofit was founded by Meta CEO Mark Zuckerberg and his wife, Dr. Priscilla Chan.
AI Models for Biological Discovery
The initiative, known as the Virtual Biology Initiative, aims to measure how cells respond to changes across far more conditions than scientists have previously studied.
The collected data will be used to create predictive AI models that could reduce the time required for drug development, a process that currently takes years.
“Biology has been just sort of a clever discovery-based science until this point,” Chan said, explaining that the project is designed as a shared scientific resource that can continue growing over time.
Biohub said the datasets will eventually be released publicly, although commercial organizations providing funding will receive early access before the information becomes widely available.
Building a Large-Scale Biology Dataset
Alex Rives, Biohub’s head of science, said current biological datasets contain information from hundreds of millions of cells, but accurate predictive models will require billions and eventually trillions of data points.
Biohub’s goal is to create a more complete understanding of biological systems by collecting and organizing data at an unprecedented scale.
“We need to capture the language of biology, we need to capture the language of the cell. And that doesn’t exist today,” Rives said.
The data will be generated through technologies such as spatial transcriptomics, which maps molecular activity inside intact tissue, and experiments that record how cells react to environmental changes.
Five-Year AI Biology Goal
Rives said the work would normally require decades to complete, but the partners aim to accelerate the process into five years.
The first dataset is expected within about one year, while Biohub expects accurate predictive models to become possible within five years.
The organization also plans to seek additional support from pharmaceutical companies and philanthropic groups to expand the project.
Growing AI Competition in Healthcare
The initiative comes as major AI companies increase investment in biological research and healthcare applications.
Anthropic has expanded its biology efforts through a dedicated laboratory, while the OpenAI Foundation has launched a grant program of more than $125 million to support biological and medical datasets for AI research.
