Stanford's 37,000-agent virtual biotech: product lessons beyond drug discovery

Stanford Medicine published a Science study on a virtual biotech company of tens of thousands of AI agents covering drug development. The system identified success signals for candidates and proposed an ADC design against B7-H3 using data up to January 2025; a pharma company later independently reached a similar strategy that received FDA breakthrough therapy status.

• A system of 37,000 AI agents covers the full drug development cycle
• Agents proposed an ADC design against B7-H3 on data up to January 2025
• A pharma company independently reached a similar strategy with FDA breakthrough status
• Key lessons: role specialization, shared artifacts and external validation