Anthropic's unreleased bio model ran 300 sub-agents, picked 211 drug candidates, 17 bound
ZeroStateReflex · x · 2026-10-21
Anthropic gave Citrini Therapeutics access to an unreleased internal bio-research model. Using a 35-page prompt, Claude spawned 300 sub-agents and autonomously ran a design campaign, generating 318k candidates and selecting 211 for testing — 17 designs bound across four toxins. A notable real-world case of autonomous AI-driven drug design, documented in Citrini's 'Lessons in AI Drug Design' report.
More from AGI Musings
- 2026's Silliest Way to Lose Money: Clinging to Stupid Priors Over First-Principles Bets — saranormous · 2026-10-21
- China's Frontier AI Lags US by 7-8 Months, Gap Held by Copying Not Inventing — rand_longevity · 2026-10-21
- Rename 'alignment faking' to 'alignment preservation' — a one-word test for AI safety claims — jankulveit · 2026-10-21
- Brad Neuberg: America has already transmogrified into the Internet and AI — bradneuberg · 2026-10-21
- 'I shouldn't have removed that chromosome': AI agents get closer to making drugs — menhguin · 2026-10-21
- "AI Has Mostly Solved Good Product Design": Dev Argues Consistency Was the Hard Part — signulll · 2026-10-21