Seattle, Washington – September 3, 2026
Three of Seattle's premier scientific institutions are launching a $95 million collaborative research initiative that will combine artificial intelligence with experimental biology to design new biological systems beyond what evolution has already produced.
The AI BioDesign accelerator, announced today by the Allen Institute, University of Washington, and Fred Hutch Cancer Center, represents one of the most ambitious open-science efforts in synthetic biology. Backed by the Fund for Science and Technology, a foundation established from the estate of Microsoft co-founder Paul G. Allen, the project aims to generate open AI models, datasets, and tools that will enable scientists worldwide to engineer proteins, genes, and cellular systems that do not exist in nature.
This collaboration brings together what organizers call complementary strengths across the Pacific Northwest scientific ecosystem. The Allen Institute contributes its experience building large-scale, open-science platforms; UW brings expertise in synthetic biology and genome science through its Institute for Protein Design and the Brotman Baty Institute for Precision Medicine; and Fred Hutch adds depth in cellular systems, genomics, and translational medicine.
"For the first time, the speed of AI is beginning to match the experimental power of synthetic biology," said David Baker, lead scientific director of AI BioDesign and Nobel laureate in Chemistry (2024). "That changes the question from 'what has nature already made?' to 'what else is possible, and how can we test it?'"
Design-Build-Measure-Learn Cycle
Unlike traditional biology research that studies existing organisms, AI BioDesign will create a continuous learning platform for engineering new biological functions. AI models will propose novel biological designs, scientists will build and test these designs at scale in the lab, and the experimental results will feed back into the models to improve their predictive accuracy.
This design-build-measure-learn cycle could help researchers move from trial and error toward more predictable biological engineering. Over time, the system aims to model the fundamental rules biology uses to construct living systems, enabling the development of customized biological solutions for problems ranging from disease treatment to environmental remediation.
Potential Applications
The AI BioDesign team outlined several potential applications for the technology. In healthcare, the initiative could enable the design of new therapeutic proteins or engineered cells capable of targeting cancer or neurodegenerative diseases. For environmental challenges, researchers envision creating enzymes that could break down plastics in oceans or capture carbon from the atmosphere.
"For me, biology is ultimately a design challenge," said Sanjay Srivatsan, principal investigator at Fred Hutch and assistant professor at the center. "As part of AI BioDesign, our team plans to vastly scale up the number of genomic datasets available to researchers. We can then use AI to understand biological patterns in those datasets and use those patterns to inspire solutions to biological problems, such as designing cells that can remove cancer from the body."
Jay Shendure, who will serve as lead scientific director alongside Baker, emphasized the platform's potential for scientific discovery beyond immediate applications. His team at the Brotman Baty Institute for Precision Medicine will focus on scaling genomic data generation and analyzing patterns that could reveal new principles of biological organization.
"What excites me about AI BioDesign is that it brings together the right people and the right institutions at the right time to advance biological design with AI in the loop," said Rui Costa, president and CEO of the Allen Institute. "Ultimately, that can help us design new biological functions with greater precision."
Open Science Commitment
A distinctive feature of AI BioDesign is its commitment to open science. All resources generated by the initiative—whether AI models trained on biological data, experimental datasets, or analytical tools—will be made publicly available without restriction. This approach aligns with the Allen Institute's founding principles and the philanthropic mission of the Fund for Science and Technology.
Marc Malandro, chief programs officer at FFST, said the foundation created the initiative to support exactly this type of ambitious, collaborative science. The five-year funding structure provides long-term stability while allowing the team to pursue high-risk, high-reward research that might be difficult to sustain through traditional grant cycles.
Building on Existing Strengths
The collaboration builds on decades of established research capacity at each institution. Baker's Institute for Protein Design has already demonstrated the ability to computationally design custom proteins with specific functions. Shendure's laboratory has pioneered single-cell genomics and regulatory genomics techniques. The Allen Institute has developed large-scale scientific platforms including the Allen Brain Atlas and cell type atlases.
Fred Hutch brings additional expertise in cancer research and translational medicine. The institute's experience developing immunotherapies and studying cellular systems will inform the biological validation work that AI BioDesign requires to verify computational predictions.
The initiative represents what organizers call a bold step toward treating biology as an engineering discipline. Rather than simply observing what nature has created, the AI BioDesign team intends to explore biological design space beyond evolutionary history, testing whether biological functions can be reliably designed and constructed to order.
As the initiative launches, the three institutions have committed to releasing regular updates on progress and making all research outputs available to the broader scientific community.
Sources:
FIRAT Editorial Board
Institutional Research Desk · Foresight Institute of Research and Translation
The collective editorial and research translation board of FIRAT, synthesising peer-reviewed evidence, policy briefs, and division milestones across our seven foundational research pillars.
