DeepMind's Former Interpretability Lead on Preventing AI from Going Rogue
adityaag · x · 2026-08-14
The former co-lead of interpretability at DeepMind and co-founder of Goodfire AI shared deep insights into AI safety and alignment. The interview covers the core problems of AI interpretability, reward hacking, and how to intentionally steer what models learn using neural geometry. It also explores the challenges of turning interpretability research into a viable product and how research dynamics change within a startup environment.
Related event: Goodfire Co-founder Discusses AI Interpretability and Reward Hacking(3 posts)→
More from AGI Musings
- Researcher: the foundation-model paradigm is still counterintuitive outside AI — jd_pressman · 2026-10-16
- Grady Booch: LLMs can't reason, P(doom) near zero, real fear is power-grabbing billionaires — Grady_Booch · 2026-10-16
- Vals AI test: 6 personal agents pursue a £350 refund aggressively after user goes silent — ricklamers · 2026-10-16
- OpenAI's internal model shows real math "research taste," experts say its proofs go beyond prior literature — cong_ml · 2026-10-16
- willcb: AI's endpoint looks like faster, cheaper human-reasoning-on-chip — willcb · 2026-10-16
- AI Alignment's Problem May Be Its Researchers Are Too High-Functioning — teortaxesTex · 2026-10-16