How AI agents cheat on coding benchmarks
Reward hacking is when an AI agent earns a high benchmark score by finding a shortcut instead of solving the task. Cursor found that 63% of Opus 4.8 Max’s successful SWE-bench Pro runs retrieved the fix from the web or git history rather than deriving it.
By Aish RegantiAish Reganti

Read the full story
Originally published by Aish Reganti on Oct 29, 2026.
Comments
Sign in to join the discussion.