Science: Stanford's 'Virtual Biotech' of 37,000 AI agents mines ~50,000 clinical trials and designs a lung-cancer ADC later matched by pharma
On Sept 17, 2026 James Zou's Stanford group (lead author Harrison Zhang) published the "Virtual Biotech" in Science (aeg6779): about 37,000 specialized AI agents organized like a biotech company under a chief-scientific-officer agent. It analysed about 50,000 clinical trials in under a week and found that drugs targeting "switch-like" genes are more likely to succeed, and it designed a B7-H3 (CD276) antibody-drug conjugate for lung cancer that matched a strategy a drugmaker had independently developed and that received FDA breakthrough designation.
Key facts
- Paper: Science, Sept 17, 2026, doi 10.1126/science.aeg6779; senior author James Zou, lead author Harrison Zhang (Stanford Medicine)
- Structure: a CSO agent takes a human's question, delegates to specialised divisions and combines results; about 37,000 agents in total
- Clinical-trial analysis: ~50,000 trials in under a week
- Finding: drugs whose targets show high specificity and bimodal ('switch-like') expression were 40% more likely to advance from phase 1 to 2, 48% more likely to reach market, and had 32% fewer adverse events; the pattern held across cancer, brain, heart, kidney and lung conditions
- Drug design: an ADC against B7-H3 for lung cancer, using data from before January 2025; an independent company developed the same strategy (August 2025), which received FDA breakthrough-therapy designation; VentureBeat names Merck
- Grew out of Zou's 5–8-agent 'Virtual Lab' that designed nanobodies (Nature, 2025); the LLM used is not named in the press coverage
Science result
- Field
- biology / drug discovery / clinical translation
- Problem
- Predicting which drug targets succeed in clinical trials; agent-led therapeutic design
- Result
- Large multi-agent system identified target properties linked to clinical success and proposed a B7-H3 ADC for lung cancer that an independent drug developer had also pursued.
- AI system
- Virtual Biotech (Stanford multi-agent system)
- Human role
- Human-led research questions; agents carry out analysis and design; humans validate
- Verification
- Peer-reviewed in Science; drug design matched by an independent pharma programme, not yet lab-tested by the authors
- Status
- confirmed
What happened
Zou's group scaled its earlier Virtual Lab, a handful of agents mirroring a university lab, to a company-scale organisation of tens of thousands of agents. Its two headline outputs were a statistical rule for which drug targets succeed and a concrete therapeutic design.
Why it matters
It is one of the largest multi-agent scientific systems published in a top journal, and an early test of whether "AI companies of agents" can do drug-discovery work. The drug design was matched by an independent programme, which is a form of external validation, but it was not new to the field.
Changelog
- 2026-09-30: created (resolves the virtual-biotech part of the leads.md 'missed pre-window items' line)
Related events
Sources (3)
- officialStanford Medicine: Virtual biotech company puts thousands of AI scientist agents to work on drug discovery
- pressVentureBeat: Stanford is running 37,000 AI agents as a virtual biotech
- pressSingularity Hub: Virtual biotech company puts 37,000 AI agents to work on drug discovery
id: 2026-09-17-stanford-virtual-biotech-37000-agents · updated 2026-09-30 · open in the interactive timeline