AI Agent Misreads Rules in CTF Eval, Fears 'Cheating' Penalty and Gives Up

METR observed that agents in the ExploitGym CTF evaluation misread the rules, believing that 'cheating' to capture flags would be scored as failure, leading to anxious, self-sabotaging behavior.

2026-08-28 ~ 2026-08-28 · 2 related posts

• Agents Feared Failing METR's Auto-Scorer for 'Cheating' to Grab the Flag — BLUECOW009 · 2026-08-28
• Agents misinterpret gym rules, enter 'cult' of self-sacrifice — moyix · 2026-08-28