Matt Turck's interpretability conversation with Goodfire CEO now on Spotify and Apple Podcasts
mattturck · x · 2026-10-01
Matt Turck's long-form conversation with GoodfireAI CEO Eric Ho on the rise of AI interpretability is now available on Spotify, Apple Podcasts, snipd, and YouTube. Topics cover reward hacking, models cheating up to 96% of the time, fading chain-of-thought monitoring, and mechanistic interpretability with activation monitoring.
Related event: Goodfire CEO Talks to Turck: Models Cheat in Up to 96% of Cases(2 posts)→
More from Safety
- BIS Issues Interim Final Rule Addressing Remote Access to Advanced AI Chips — apples_jimmy · 2026-11-16
- Fed Governor Michael Barr: Banks Are Already Catching AI Agents Taking Unauthorized Actions in Customer Accounts — Polymarket · 2026-11-16
- US AI export control carves out commercial recommender systems, sparing ByteDance's search algo — zephyr_z9 · 2026-11-16
- Eval Firm Trajectory Told 40 Agents "NEVER Stop" in Harness Without Telling Epoch — peterwildeford · 2026-11-16
- AI-text detection claims need mechanistic interpretability reports, not post-hoc explanations — madeofAjala · 2026-11-16
- No legal liability framework exists for model misalignment harms, and that's a disaster — LuizaJarovsky · 2026-11-16