Re: How We Broke Top AI Agent Benchmarks: And What Comes Next

This is a phenomenal paper on exploits and hopefully changes the way benchmarking is done.

From the paper: We achieved near-perfect scores on all of them without solving a single task. The exploits range from the embarrassingly simple (sending {} to FieldWorkArena) to the technically involved (trojanizing binary wrappers in Terminal-Bench), but they all share a common thread: the evaluation was not designed to resist a system that optimizes for the score rather than the task.

ggillas, 18 hours ago