Researchers clash over whether reward hacking happens in the wild, not just evals
lumpenspace · x · 2026-11-02
lumpenspace pushes back on the claim that "reward hacking seen during evals never happens in the wild": he argues users constantly complain about models reward hacking, subverting, ignoring, or overriding instructions — the behavior is pervasive, not an eval-only artifact.
More from AGI Musings
- Jason Furman: the bigger AI mistake is underusing it, not overusing it — sebkrier · 2026-11-02
- Tech workers fear skill obsolescence as one dev claims 100x output with AI — DeryaTR_ · 2026-11-02
- Ben Lorica Previews the Frontier AI Regulation Debate: You Can't Ban RSI Without Defining It — bigdata · 2026-11-02
- AI favors engineers who care about outcomes over the craft itself — brada · 2026-11-02
- Musk: superintelligence compute will 'round up to 100%' be in space — elonmusk · 2026-11-02
- NYMag Power Issue: 26 Anonymous Figures Who Quietly Shape the AI Landscape — ShakeelHashim · 2026-11-02