Biohub is leading a $1.8 billion initiative to build artificial intelligence models capable of predicting cell behavior to speed up drug development. The nonprofit, backed by Mark Zuckerberg and Priscilla Chan, is coordinating the massive effort spanning data, lab equipment, and compute. The collaboration involves contributions from major tech companies and government agencies, building upon a prior $500 million pledge made in April for the five-year Virtual Biology Initiative.
As part of the funding, Meta, Google DeepMind, and Isomorphic Labs are contributing a combined $300 million, while the US Department of Energy is investing over $500 million in lab measurements and compute resources over five years. Additionally, the National Institutes of Health are coordinating datasets built with more than $500 million in federal funding, which Biohub will standardize for AI training purposes. Commercial funders will receive one year of exclusive access to the resulting data before it is made public, whereas government-funded work will be unrestricted.
This project joins a broader industry trend of artificial intelligence companies entering the biological research space. Anthropic has previously established its own biology lab focused on AI-driven drug development, and the OpenAI Foundation is dedicating over $125 million toward biological and medical datasets. A first dataset from the Biohub-led initiative is expected to be ready in about a year.
Why it matters: This unprecedented public-private collaboration unites massive compute and standardized datasets to accelerate AI-driven discoveries in biology and medicine.