Isomorphic Labs President on Using AI to Turn Drug Discovery into Precision Engineering
No Priors · youtube · 2026-11-12
On the No Priors podcast, Elad Gil talks with Isomorphic Labs president Max Jaderberg about AI-driven drug discovery:
- Core thesis: move discovery onto computers, turning pharma from empirical trial-and-error into precision engineering; AlphaFold was the first proof that deep learning can generalize in biology.
- Compounding at scale: AI models compound learnings to gain generalizability, raising success probability and speed in real-world clinical trials.
- Beyond animal testing: in-silico models could eventually replace animal experiments.
- Platform approach: building drug design engines agnostic to disease area, target class, and modality; accurate molecule engineering unlocks multidrug cocktails spanning complex biological pathways.
- Long-term goal: AI molecule optimization to improve both human lifespan and healthspan.
Also covers hiring, datasets, clinical endpoint design, and internal development vs. partnerships.
More from AGI Musings
- OpenAI Foundation hires Head of AI for Alzheimer's, plans $1B first-year investment — MichaelRetchin · 2026-11-12
- Over-relying on AI for life decisions: one user's costly lessons — Wise_Breadfruit7932 · 2026-11-12
- Sabine Hossenfelder argues the economics of AI don't add up — skdh · 2026-11-12
- Dev argues LLMs can't yet do out-of-distribution science like novel algebraic geometry — burny_tech · 2026-11-12
- Jeremy Howard: AI beating humans at nerdy tasks isn't a strong indicator of ASI — jeremyphoward · 2026-11-12
- Michael Tomasello: lying only works because humans are cooperative by default — Machine Learning Street Talk · 2026-11-12