Google DeepMind, Isomorphic and Meta put $300 million into Biohub's open data for a virtual cell

The labs are paying for open data, and NIH is contributing datasets it already paid for.

By Think Facility · October 7, 2026 · Google DeepMind Research

On October 7, 2026, Biohub announced that Google DeepMind, Isomorphic Labs and Meta are putting a combined $300 million into its Virtual Biology Initiative. The US Department of Energy and the National Institutes of Health signed on the same day, and Biohub puts the whole package at $1.8 billion in funding, data, computation and measurement technology.

The money buys data. The goal is an open resource big enough to train AI models that predict how a cell responds to a drug or another intervention, what researchers call a virtual cell. Biohub calls it the largest coordinated commitment to AI-ready biological data to date.

Total: $1.8 billion in funding, data, computation and technology
DeepMind, Isomorphic, Meta: $300 million, collectively
DOE: more than $500 million over five years
NIH: datasets from more than $500 million in prior federal investment
Biohub: founding $500 million commitment, announced April 2026

Who is paying for what

The Department of Energy's share is the biggest new cash: more than $500 million over five years for lab measurement, modeling and computing, run through its Genesis Mission and the national labs' supercomputers and microscopes.

Darío Gil, smiling, in dark-rimmed glasses, a gray suit and a red tie, in front of an American flag and a blue backdrop — Darío Gil, the Department of Energy's Under Secretary for Science, in his official DOE portrait. Photo: U.S. Department of Energy, Public domain, via Wikimedia Commons

NIH's part works differently. It isn't adding fresh money in the release. It's bringing datasets and repositories built with more than $500 million of earlier federal spending, and Biohub will put them into a standard form for training models.

The three companies' $300 million is a joint figure. The release doesn't split it between them, so there's no way yet to say how much is Meta's (that's the number I'd most like to see).

What the labs said

Demis Hassabis posted about the deal from his own account a little before noon Eastern. His post names CZI, the Chan Zuckerberg Initiative, as the partner.

Demis Hassabis@demishassabis · GoogleOct 7, 11:53 am ET

The most important thing we can apply AI to is improving medicine and human health - very excited about this partnership with the CZI and the Virtual Biology Initiative!

View the post on X

Inside the release, Google DeepMind's Pushmeet Kohli made the case for why the data has to be shared.

We will not solve this challenge without open, experimental biological data at an unprecedented scale, showing how living cells behave and respond to changes

From AI-ready biological data: $1.8 billion global commitment

Max Jaderberg, president of Isomorphic Labs, said his company joined as a founding member. Meta isn't quoted anywhere in the release, and I didn't find a post from Meta or its executives about its share.

Who else is in

Biohub lists the Allen Institute, Broad Institute, Gladstone Institutes, the Human Cell Atlas, the Human Protein Atlas and the Wellcome Sanger Institute as partners helping organize the science. NVIDIA is supporting the computing side, and Renaissance Philanthropy is helping raise more money for data generation.

Biohub says it's building shared standards, common identifiers and a single point of access, so data from all these groups can be used together. The DOE money is spread over five years, and the release gives no date for the first open datasets.

More on Google

All Google stories