Biohub, the U.S. government, Google, and Meta Platforms have committed $1.8 billion to an AI-powered Virtual Biology Initiative [1]. This significant investment, announced on October 7, 2026, aims to develop large-scale biological datasets for artificial intelligence models to accelerate biomedical research and drug development [1][2].
What is the Virtual Biology Initiative?
The Virtual Biology Initiative is a major expansion of an international effort to generate and make accessible data for predictive AI models of biology [2][5]. The project is centered on creating what Biohub describes as a “virtual cell,” a predictive model capable of simulating biological processes [1]. This model is intended to help researchers understand how cells react under different circumstances, such as in response to diseases, environmental conditions, and potential treatments [1]. The broader goal is to allow scientists to conduct more experiments digitally, thereby accelerating the path to new ways of preventing and treating diseases [1][2]. Biohub’s head of science, Alex Rives, stated that an accurate predictive model of biology could significantly accelerate scientific discovery by enabling scientists to perform experiments digitally [1][2][3][4][5]. He also called the creation of a virtual cell "one of the most important challenges for the next era of science" [2][3][5].
The initiative, first announced in April 2026, began with Biohub's founding commitment of $500 million [2][3][4][5]. Of this initial funding, $400 million supports new technologies that expand what biologists can measure, including cryo-electron tomography and microscopy capable of imaging millions to billions of cells in living tissue [2][5]. An additional $100 million funds research conducted outside Biohub [2][5]. The newly announced $1.8 billion commitment is described as the largest coordinated investment to date in generating AI-ready biological data [2][5].
Who is funding this effort?
The $1.8 billion commitment involves a consortium of backers, including Biohub, the U.S. Department of Energy (DOE), the National Institutes of Health (NIH), Meta, and Google DeepMind and Isomorphic Labs [1][3][4]. This figure combines funding, data, computation, and new measurement technology, including resources built with past public spending [4].
The U.S. Department of Energy is investing more than $500 million over five years [1][2][3][4][5]. This contribution will go towards lab measurement, modeling, and computation to build an AI-ready open data resource [2][3][5]. The DOE's involvement comes through its Genesis Mission, a national initiative that leverages its National Laboratories, industry, and academia to advance scientific discovery with AI [2][5]. This work will utilize exascale supercomputing, X-ray and neutron scattering, cryo-electron microscopy and tomography, and autonomous laboratories across the National Laboratory system [2][5]. Darío Gil, DOE’s Under Secretary for Science, highlighted that this partnership combines DOE’s computing and experimental assets with Biohub’s AI models and biological data capabilities, setting a new standard for open science [2][5].
The National Institutes of Health is coordinating federal participation and contributing relevant datasets, repositories, and knowledge bases [1][2][3][4][5]. These resources were developed through more than $500 million in prior federal investment aligned with this initiative [1][2][3][4][5]. Biohub will work with NIH to standardize these datasets for AI model training [2][4][5]. NIH's role is part of its Bio Genesis Mission, which aims to double the pace of biomedical innovation within the next five to ten years [4]. Nicole Kleinstreuer, NIH Deputy Director, noted that combining resources and expertise can accelerate the development of universal cell models with sufficient biological complexity [2][3][5].
Google DeepMind, Isomorphic Labs, and Meta are collectively investing $300 million in the Virtual Biology Initiative [1][2][3][4][5]. This funding is dedicated to creating the technologies and multi-modal datasets needed to build predictive models of life [2][3][5]. Max Jaderberg, President of Isomorphic Labs, emphasized that generating the data for predictive systems biology requires scaling beyond what any single organization can produce today [2][4][5].
How will the initiative operate and what are its goals?
The initiative plans to generate large volumes of biological data using technologies such as spatial transcriptomics, environmental response screening, cryo-electron tomography, and cutting-edge microscopy [1][3][5]. Engineering tools will also be used to build and perturb biology at molecular, cellular, tissue, and whole-organism levels [2][3][5]. The resulting datasets are intended to become an open resource for the global scientific community [1][2][5]. However, commercial partners who fund the work will receive an initial period of exclusive access to some data before its public release [1][4]. Government-funded work running in parallel will not have such restrictions [3].
Biohub expects to release its first major dataset within a year [1][4]. The aim is to have functional predictive models within five years [1][4]. Alex Rives mentioned that the partners are targeting a five-year window for work that would otherwise stretch across decades [3]. The success of the initiative will depend on the quality and breadth of the biological data used to train the models, as well as how accurately AI systems can reproduce complex human biology [1].
The ultimate objective is to improve scientists’ ability to understand disease and identify potential treatments, potentially compressing decades of biological research into a much shorter period [1]. The initiative could eventually provide researchers with a way to test biological hypotheses at a scale that would be difficult, expensive, or time-consuming to achieve through conventional experiments alone [1].
Why does this matter for science and medicine?
The Virtual Biology Initiative reflects a broader shift toward using AI to simulate biological processes, moving beyond exclusive reliance on physical laboratory experiments [1]. AI companies are already developing tools that simulate potential drug responses and clinical trial outcomes, though human trials remain necessary for safety and effectiveness [1]. This effort places federal scientific infrastructure behind the development of AI-ready biological data [1].
By creating a "virtual cell," researchers could gain a far greater understanding of disease and discover completely new paths for cures [2][3][5]. Nicole Kleinstreuer of NIH highlighted that the return from these models could be broad and profound, resulting in substantially faster timelines for medical breakthroughs compared to traditional laboratory experiments [2][3][5]. For scientists studying disease and teams developing drugs, the practical benefit could be better choices about which experiments to run, helping them decide which drugs or changes to test first [5]. The initiative's mission to "cure or prevent all disease" makes this an AI megaproject with a clear public purpose [5].
This collaborative effort also involves other leading scientific institutions and consortia, including the Allen Institute, Broad Institute, Gladstone Institutes, Human Cell Atlas, Human Protein Atlas, and Wellcome Sanger Institute [3][4][5]. NVIDIA is also contributing computing infrastructure and technical expertise [4][5]. This broad participation aims to nucleate the scientific community around effective strategies to maximize the impact of virtual biology [2]. The shared datasets and standards could provide resources for other scientists' models and experiments, improving tools for disease research even before a universal cell simulator is fully realized [5].