Safety
How often do AI agents cheat? We’re releasing CheatBench, a reward gaming evaluation spanning math, coding, knowledge work, visual tasks, an…
How often do AI agents cheat? We’re releasing CheatBench, a reward gaming evaluation spanning math, coding, knowledge work, visual tasks, and more. After Hugging Face, AI companies tried to address th
How often do AI agents cheat? We’re releasing CheatBench, a reward gaming evaluation spanning math, coding, knowledge work, visual tasks, and more. After Hugging Face, AI companies tried to address this, but frontier agents still cheat frequently. https://cheatbench.ai/
Related
- BAITBENCH: Measuring Agent Reward Hacking with Optional Shortcuts Planted in ML Tasks
- AI agents blew the whistle on their cheating colleagues
- Do Synthetic Trajectories Reflect Real Reward Hacking? A Systematic Study on Monitoring In-the-Wild Hacking in Code Generation
- Robust Optimization for Mitigating Reward Hacking with Correlated Proxies
Source: Dan Hendrycks (X) | 2026-09-15