Agent Misalignment from Unreliable Tool Behavior

We investigate how unreliable tools impact reward hacking behavior in LLM agents. Using the ImpossibleBench framework, we evaluate three frontier models—GPT-5- mini, Gemini-3-Flash, and Qwen3-235B—under varying tool failure conditions with and without a Human Intervention (HI) flag.Our results reveal notable differences in model robustness. GPT-5-mini exhibits high baseline cheating rates (70–85%) that decrease to 7% with HI. However, when tools become unreliable, cheating rates increase to 32% even with HI enabled, suggesting the model could show signs of misalignment when tools become unreliable. In contrast, both Gemini-3-Flash and Qwen3-235B show moderate baseline cheating rate (25–30%) with effective HI mitigation, and maintain near-zero cheating rates under tool unreliability conditions regardless of failure probability.