Polyphron, a company building artificial human biology to transform drug development and treat disease, announced $20M in seed financing led by Quiet Capital, with participation from Gradient, Haystack, and Compound. Polyphron trains frontier AI models to manufacture and simulate living tissue to create the verification substrate for AI-driven biology.

AI systems are becoming increasingly capable of generating drug candidates, biological hypotheses, and experimental designs. But the ability to generate ideas is advancing much faster than the ability to test them in human biology. Existing validation systems force a tradeoff between biological relevance and speed: Clinical trials and animal studies are slow and expensive, while simpler in vitro and computational systems fail to capture many of the multicellular dynamics that determine human response.

Polyphron combines AI-driven tissue manufacturing, stem-cell biology, and autonomous experimental infrastructure with tissue world models trained on longitudinal data from living tissues. Together, the physical and simulated systems create a verification substrate for human biology: AI systems can generate predictions, test them in donor-diverse human tissue, learn from the resulting trajectories, and improve over time. In practice, this could mean predicting whether a new drug will cause cardiac or liver toxicity before it reaches the clinic, or identifying which patients are most likely to respond to a therapeutic intervention.

Polyphron’s first tissue is cardiac, now in production in the company’s tissue foundry across multiple donor lines, with a broader donor panel being onboarded to capture human biological variation. Liver follows this year, extending the same manufacturing, automation, and longitudinal measurement stack to drug metabolism and liver toxicity.

“AI will not cure all diseases without a verification layer. The cost of producing another new therapeutic hypothesis or intervention in many modalities could soon approach zero but this will not lead to cures until something can answer, ‘What will this intervention actually do to a heterogeneous human biological system over time?’” said Matthew Osman, Polyphron’s co-founder and CEO. “The missing piece is a scalable environment where models can test predictions against living human biology – and our bet is that tissue is the right abstraction layer for this scalable environment.”

Polyphron’s Tissue Testing Environments combine physical and simulated human biology into a closed-loop system for systematically interrogating how human tissue behaves across interventions and genetic backgrounds. They comprise two components:

  • The Tissue Foundry: a physical, automated system that grows human micro-tissue from stem cells using the same developmental signals the body uses to build tissue during development

  • The Tissue World Model: a simulation layer trained on tissue trajectories to predict how tissue responds to genetic, chemical, biological, and environmental changes across donors

The long-term objective is to model human response before a clinical trial is run: predict how an intervention will behave across biological diversity, identify who is most likely to respond, and ultimately search for interventions that move biology toward a desired state.

“We back founders who are building infrastructure that changes what’s economically possible, not just what’s scientifically interesting,” said Michael Bloch of Quiet Capital. “If you can make tissue a manufacturable, testable unit at scale, you change the economics of drug development. Polyphron is applying that infrastructure mindset to human biology.”

The company was co-founded by Osman and Fabio Boniolo, Ph.D., co-founder and Chief Scientific Officer, previously at Dana-Farber Cancer Institute, the Broad Institute, and Harvard Medical School. Earlier this month Vinh Q. Tran, previously a post-training and self-improvement researcher for Gemini at Google DeepMind, joined as an additional co-founder and Chief AI Scientist. The team also includes George Pilitsis, Chief Operating Officer, who founded and scaled the Datapoints product line at Ginkgo Bioworks.

About Polyphron

Polyphron is building artificial human biology. The company combines automated tissue manufacturing, and AI simulation to create a physical-and-simulated verification substrate for therapeutic hypotheses against living human biology. Polyphron is based in New York. For more information, visit www.polyphron.com.

Media gallery

About The Author
Business Wire