In 2016, the Chan Zuckerberg Initiative launched with a mission to support the science and technology to make it possible to cure, prevent or manage all diseases by the end of this century. In the years since, our multidisciplinary teams of scientists and engineers have built incredible technologies to observe, measure and program biology. We’ve created groundbreaking, large-scale data resources for cellular biology and molecular imaging that have led to new insights about human biology, and developed state-of-the-art models and scientific tools to accelerate the work of biologists worldwide.

Our work centers on four core scientific grand challenges: developing a unified, AI-based model of the cell to predict and understand how cells behave within the human body; advancing state-of-the-art imaging systems to visualize complex biological processes at unprecedented scales; creating new instrumentation to monitor and modulate inflammation in real time; and using AI to reprogram and harness the immune system for early detection, prevention, and treatment of disease. Together, these efforts aim to decode the inner workings of human biology and enable scientific and medical breakthroughs.

To tackle these challenges, we’ve built partnerships with the scientific community and consortia that have helped to shape the scientific agenda over the last decade. This includes our Billion Cells Project, a landmark single-cell sequencing project to advance researchers’ understanding of cellular behavior and gene function. We’ve pursued our work with a commitment to open science.

Today, the extraordinary progress in artificial intelligence is beginning to open up new frontiers. One of the most important impacts that AI is likely to have for society is to accelerate progress toward solving disease. Yet we still lack a fundamental understanding of the biological systems that underlie health and disease. Trillions of cells — each built from millions of molecular components programmed by the genome — interact in ways far too complex for current technologies to model or simulate.

AI is beginning to change how we understand biology. It is already enabling the design of protein molecules and starting to make biology programmable, much like engineering machines or coding software. But understanding how individual cells and systems of cells drive health and disease remains an immense challenge. Solving this challenge will open up new frontiers for medicine. This is a problem that cannot be solved by computation alone — it will require the development of innovative new technologies to generate multiparametric data for cellular biology at scale, and innovative new artificial intelligence approaches that can learn from these data to build a unified model of biology that captures the full complexity of the cell, and eventually human physiology.

Today, we are launching a large-scale scientific initiative to push the frontier of artificial intelligence in biology.

We believe this is a pivotal moment in science, and the future of AI-powered scientific discovery is starting to come into view. We are putting the resources behind this effort that we believe will be needed to achieve this: frontier-level compute power, frontier technologies for instrumenting, imaging, and programming biology, and exceptional biological and AI science. To power this initiative, we are uniting our scientific teams as a single organization that will be known as Biohub. This name reflects both the history of our organization and our vision for the future.

The team at EvolutionaryScale, a frontier AI research lab and public benefit company that has created groundbreaking, large-scale AI systems for the life sciences, will join Biohub to help advance this initiative. Alex Rives, EvolutionaryScale’s co-founder and chief scientist, and a pioneering scientist in artificial intelligence for biology, will serve as head of science, leading an integrated research program across experimental biology, data, and artificial intelligence. The EvolutionaryScale researchers will join Biohub’s team of leading scientists and technologists. This combined team of biological scientists, engineers and AI scientists will work to develop the datasets, experimental technologies, and modeling innovations needed to build the next generation of models to power the future of biology. To support this initiative, we continue to expand our compute capacity, which will grow to 10,000 GPUs by 2028.

We believe that it will be possible in the next few years to create powerful AI systems that can reason about and represent biology to accelerate science. Here’s what we think this will look like:

• Virtual biology: As we build high-fidelity digital representations of molecules, genomes, cells, and living systems, we’ll increasingly be able to ask scientific questions digitally — conducting virtual experiments at a scale and pace far beyond what’s possible in the lab. Digital representations of biology will enable scientists to simulate experimental outcomes and explore biological systems with unprecedented speed and depth, allowing them to ask and answer fundamental questions in ways never before possible.
• Accelerated discovery: We are simultaneously working toward scientific artificial intelligence that can reason about, learn from, and synthesize the world’s scientific data, to reveal fundamental insights and accelerate the rate at which new scientific discoveries can be made.

We will research, build, and scale up these powerful technologies — and make them available to scientists to drive basic advances in biology and frontier medical applications. As we make progress on these kinds of systems, we believe it might eventually become possible to achieve decades of discoveries in months.

We believe that this will come together to unlock frontier medicine. AI could enable advances like early disease detection, monitoring, and prevention; programmable cellular medicines; personalized gene-editing-based cures; and approaches to prevent and cure inflammatory and chronic disease.

As we build toward this longer-term roadmap, today we are also sharing several updates on our work in progress:

We are launching the Virtual Immune System project, a flagship effort to model the immense complexity of the human immune system. We aim to change the understanding of immunological science — opening the door to engineering human health, simulating immune therapies, reprogramming dysfunctional cells and preventing diseases before they arise.

We’re also making three new models freely available on our virtual cells platform:

• VariantFormer: a model that directly translates personal genetic variations into tissue-specific gene activity patterns;
• CryoLens: an end-to-end, pretrained, large-scale model for cryoET providing unsupervised structural similarity analysis; and
• scLDM: a new AI model that can generate realistic single-cell data in silico at unprecedented fidelity.

These complement our growing range of state-of-the-art models, including GREmLN, a graph-aware model for understanding how gene regulatory networks govern cell behavior, and rBio, a conversational large language model designed to perform biological reasoning and make scientific knowledge more accessible. These models are steps toward virtual representations of biology that can help scientists understand and unlock the complexity of the cell and genome.

With the rapid and continuing advances in artificial intelligence, we believe we’re on the cusp of a scientific revolution in biology — as frontier artificial intelligence and virtual biology give scientists new tools to understand life at a fundamental level, and begin to accelerate the rate of scientific progress. As these powerful technologies continue to advance, we’re committed to developing them responsibly. And, as we have always done, we’ll continue to work in partnership with the scientific community, and our datasets and models will be shared to maximize their benefit to science and humanity.