AI RESEARCH

Hack-Verifiable Environments: Towards Evaluating Reward Hacking at Scale

arXiv CS.LG

ArXi:2605.20744v1 Announce Type: new Aligning autonomous agents with human intent remains a central challenge in modern AI. A key manifestation of this challenge is reward hacking, whereby agents appear successful under the evaluation signal while violating the intended objective. Reward hacking has been observed across a wide range of settings, yet methods for reliably measuring it at scale remain lacking. In this work, we