Stanford's Noah Goodman adds monitoring and 'cutting bad links' section to AI safety lecture
noahdgoodman · x · 2026-10-19
Stanford professor Noah Goodman says he added a whole new section on "cutting the bad links" plus monitoring to his AI safety lecture, after researchers including Niloofar Mire suggested covering deception and loss of monitorability, and pointed to Boaz Barak's latest AI safety course as a reference. The exchange reflects academics systematically folding frontier safety topics like model deception and monitorability failures into university curricula.
More from Companies & People
- Dev jokes: startup contest named "Cold Start" even though AI killed cold start — threepointone · 2026-10-19
- OpenCode hack prompts Suhail to warn: corporate America will almost certainly get breached too — matthew_d_green · 2026-10-19
- Every engineering leader is building the same AI automations right now — generativist · 2026-10-19
- AI will crush SaaS margins, making delivery and support the new B2B moat — rickasaurus · 2026-10-19
- Stephen Davies' 322-page open textbook "Blueprints" on large-scale software design joins ChapterPal — burkov · 2026-10-19
- Anthropic cofounders pledged 80% of equity to charity — 'trillion-dollar or zero-dollar company' — kevinroose · 2026-10-19