Lectures Explore LLM Over-Optimization and Reward Hacking

Nat Lambert recently delivered a lecture exploring AI challenges like over-optimization, reward hacking, and sycophancy. The talk reviewed fundamental concepts such as Goodhart's Law and how rubrics can be exploited, leading to distorted benchmark results.

2026-07-25 ~ 2026-07-26 · 3 related posts

• Nat Lambert’s shorter lecture links over-optimization, reward hacking, and leaderboard gaming — natolambert · 2026-07-25
• Nat Lambert reposts his lecture on over-optimization and reward hacking — natolambert · 2026-07-25
• Thom Wolf shares a shorter lecture on over-optimization, reward hacking and sycophancy — Thom_Wolf · 2026-07-26