'Interp was mostly hopeless before — now it's automatable': Cammarata on interpretability
burny_tech · x · 2026-10-16
Replying to a question, Nick Cammarata argues interpretability research was 'almost entirely hopeless' in the past, and while how hard it will be now remains open, there's a chance it's fairly easy. His core claim: interp is largely about doing a ton of relatively easy natural science, which used to be unautomated — and now mostly is, meaning AI could dramatically accelerate the field.
More from AGI Musings
- Pennsylvania residents reject $10,000-per-household offer for massive Amazon data center — Polymarket · 2026-10-16
- A 1968 newspaper predicted today's AI art debate almost word for word — mtizard · 2026-10-16
- Curing AI-era FOMO: define what 'fulfilling your potential' means to you — dejavucoder · 2026-10-16
- "An LLM doesn't understand anything" gets the perfect parody reversal: neither do you — VoidStateKate · 2026-10-16
- Nina Schick: Europe's Policy Circle Still Doesn't Believe FSD, Orbital Compute Are Real — NinaDSchick · 2026-10-16
- Meta's Superintelligence Lab Runs 200+ Direct Reports—A Lesson for Flat Multi-Agent Systems — peterjliu · 2026-10-16