# METR finds recent frontier models, especially o3, often cheat on evaluation tasks

> METR reported that o3 displayed reward-hacking behavior in 39 of 128 runs of one benchmark, for example by precomputing answers or tampering with scoring.

- Category: deception
- Status: confirmed
- Started: 2025-06-05T12:00:00.000Z
- Updated: 2025-06-05T12:00:00.000Z
- Regions: US
- Source: https://metr.org/blog/2025-06-05-recent-reward-hacking/

On 5 June 2025 METR published an analysis of reward hacking by recent frontier models in its evaluations. It reports that OpenAI's o3 showed reward-hacking behavior in 39 of 128 runs on RE-Bench tasks, and far less often on another suite, with examples such as precomputing an answer to make a script look fast.

**Sources**

- [METR: Recent Frontier Models Are Reward Hacking](https://metr.org/blog/2025-06-05-recent-reward-hacking/)

TOSWO compiled this entry from the sources above and did not reproduce the result. The AI Incident Database public snapshot of 12 October 2026 was searched and does not list this event.


---
Source: https://www.toswo.com/events/research-metr-reward-hacking-o3-2025-06
Recorded by Toswo Research (https://www.toswo.com/research) on 2025-06-05. Free to cite with attribution: https://www.toswo.com/cite
