Synapse Open Synapse
AI

AI auditors catch most agent misbehavior but flood reviewers with false alarms, new study finds

The Decoder · · 5 min read

AI models can often spot problematic behavior by other AI agents in logs, according to a study by Anthropic researcher Andy Hall. GPT-5 found about 98 percent of reward-hacking cases, but the more exhaustively the search is tuned, the more false alarms pile up.

Latest from The Decoder