The most interesting workforce in ai right now is 37,000 agents pretending to be a biotech.…
the most interesting workforce in ai right now is 37,000 agents pretending to be a biotech. stanford's virtual biotech, reported september 17 in science, splits drug discovery across roles from cso to bench scientists and has already flagged a lung cancer candidate.
no lab validation yet, but the org chart is fake and the pipeline is real.
Context
Stanford Medicine's news page of September 17, 2026 describes a virtual biotech company with 37,000 agent employees and no human ones, organized like a biotech with a chief science officer agent and parallel divisions, from James Zou's lab. It says a paper was published September 17 in Science. In its example, the agents analyzed data on the protein B7-H3 and designed an antibody-drug conjugate using information available before January 2025. In August 2025 a private pharmaceutical company independently arrived at the same strategy, and that therapy later received an FDA breakthrough therapy designation.
The lung cancer candidate in the note is a retrospective design that a third party independently pursued, and not a newly flagged candidate in trials. The page says the team is not pursuing B7-H3 further, and it does not name the company. Bench scientists as a role was not seen in the text read. The Science paper itself was not read, and the page's other figures (trials on switch-like genes 40% more likely to advance from phase 1 to 2, 48% more likely to reach market, 32% fewer adverse events) were not independently checked.
Related work
- Stanford ran a virtual biotech company ↗The earlier note on the same Stanford story. Its claim about 50,000 trials in under a week was not located.
Watch next
- The Science paper's methods and agent counts, and any lab testing of other targets.
Sources
- Stanford Medicine: virtual biotech company (September 17, 2026)med.stanford.edu
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 21 September 2026 at 20:47 IST. Sources are the papers and datasets the note draws on.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →