A biotech startup claims the AI drug-discovery industry lacks the right data to deliver on cancer cures. Vivodyne’s HIVE robotic labs grow 20 types of human tissue, autonomously dosing and monitoring them to generate causal biological data missing from current AI models. “Absent human testing, what are these models going to do?” asks CEO Andrei Georgescu. “They’re going to cure cancer in mice.”
The company opened what it calls the world’s largest “human data center” near San Francisco last week. Vivodyne says its liver tissue matches human trial toxicity data with 94% accuracy, airway tissue aligns 96%, and bone marrow shows 100% concordance across 20 chemotherapy drugs. With $80 million raised, the startup claims its throughput is twice that of all U.S. animal trials combined.
Vivodyne is working with major pharma partners to reduce clinical trial failures, comparing the approach to automotive crash testing. Georgescu argues the labs will train AI models to understand human biology’s causal pathways, critical for complex diseases and combination therapies. “Establishing causality in human biology is the basis of all of this.”
