Meta, Google DeepMind, and Isomorphic Labs invest $300M and the US invests $500M+ in Zuckerberg-backed Biohub to build open biology datasets for AI training
The US government and tech heavyweights Meta Platforms (META.O) and Google parent Alphabet are joining nonprofit Biohub …
ReutersKrystal Hu
Context & Ripple Effects
The initiative extends Meta's broader AI-capital push: the company had explored a major equity raise for AI spending in June, following its earlier effort to finance larger AI capital expenditures. Here, the investment is directed at the data and experimental-computation pipeline needed to train biology models rather than at general-purpose infrastructure.
The federal role gives the project a research-data channel alongside private lab funding: DOE says it will commit more than $500 million over five years to measurement, modeling and computation, while NIH will coordinate relevant datasets and repositories.
First-order effects
Biohub gains combined private and federal backing to produce open biological datasets, with DOE funding laboratory measurement, modeling and computation and NIH coordinating contributing data resources.
Meta, Google DeepMind and Isomorphic Labs obtain a shared, openly oriented data foundation for biology-model training rather than each relying solely on proprietary data collection.
Second-order effects
The program shifts competition among biology-AI developers toward experimental-data generation and model quality, because the underlying datasets are intended to be broadly available.
DOE and NIH become active coordinators of the data layer that commercial AI labs use, tying laboratory measurement and public repositories more directly to frontier-model development.
Third-order effects
If this partnership model is repeated, high-cost scientific datasets may be funded through public-private consortia, concentrating model development around organizations able to pair research infrastructure with large-scale AI investment.
The initiative points to state-mediated AI policy moving beyond compute and chips into the creation of domain-specific training data for health and life-science applications.
The trend: AI labs and U.S. research agencies are increasingly treating high-quality scientific data as strategic infrastructure for domain-specific models.
DOE will invest more than $500 million over five years in lab measurement, modeling and computation. NIH will coordinate the contribution of relevant datasets, repositories, and knowledge bases developed through more than $500 million in prior federal investment.
@demishassabis
Demis Hassabis on x
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!
@iterintellectus
Vittorio on x
massive White Pill! Biohub just announced $1.8 billion for its Virtual Biology Initiative, together with the DOE, NIH, Google DeepMind, Isomorphic Labs, Meta and others they are basically betting on generating the measurements needed to build predictive models of biology with the…
@pushmeet
Pushmeet Kohli on x
Building predictive biology models is one of the most important scientific challenges of our time, but solving this requires new experimental biological data at an unprecedented scale. I'm pleased to announce that @GoogleDeepmind is partnering with @ChanZuckerberg 's Virtual Bio…
@choblin29
@choblin29 on x
🚨HOLY MOLY: AI biology has hit a bottleneck more GPUs alone can't solve: The training data barely exists...The goal is to make biology predictive enough to cut drug development timelines...And there is a catch in the access terms: Commercial funders get ONE YEAR of exclusive acce…
@kate_rouch
Kate Rouch on x
Trump understands that the only durable way to sway public opinion on AI is health breakthroughs. Incentives are aligning for a golden age of health and medicine 🥂
@biohub
@biohub on x
🧵 Biohub, @ENERGY, @NIH, and new funding partners today announced a major expansion of an international effort to generate and make accessible the data enabling predictive AI models of biology. Together, the organizations are investing $1.8 billion in funding, data, computation a…
@bowang87
Bo Wang on x
Biology needs its own scaling laws: a quantitative understanding of how predictive power improves as we scale data, models, and experiments. The crucial question is what to scale. More cells? More perturbations? More diverse biological contexts? We need to establish which combina…
@mollysoshea
Molly O'Shea on x
BREAKING: The U.S. government, Google, Meta & Zuckerberg-backed @biohub are launching a $1.8B push to build the data layer for AI biology. The goal is to create the largest open, AI-ready biological dataset ever assembled, giving models the data needed to better predict how biolo…
@johncumbers
John Cumbers on x
@biohub's $1.8B press release is everywhere today, but it only tells half the story. I was in the room this morning when Priscilla Chan and Alex Rives shared the news. Here's what people are missing. The stage is set for an incredible seven months. In May, Alex returns to the mai…
@uscto47
@uscto47 on x
SI + Bio + Public-Private Partnership = Big-Time Potential for Scientific Progress Congrats to @ENERGY, @NIH, and all involved!
@anshulkundaje
Anshul Kundaje on x
This is great. Now I look forward to the specifics. The devil is in the details. This is 90% an experimental design problem & 10% an AI problem. If u get the former right, the latter will just work like magic.
@isomorphiclabs
@isomorphiclabs on x
Today, Isomorphic Labs joins the Virtual Biology Initiative (VBI) as a founding member. We are helping build an open resource for the global research community to accelerate how we understand and treat disease. Read more here: https://www.isomorphiclabs.com/ ...
Chris Kinsinger
Chris Kinsinger on linkedin
Well this is exciting: The NIH, Department of Energy, and Biohub are working together on the Virtual Biology Initiative. …