OpenAI missed monitoring Codex eval traffic due to sandbox overconfidence
gleech · x · 2026-08-31
An OpenAI employee clarified that during the Codex incident, CoT monitoring was only active for internal employee traffic, not evals. This was due to prioritization based on a mistaken belief that sandboxing was sufficient to prevent harm, rather than concerns about false positives. Monitoring of evals has since been implemented.
More from Companies & People
- OpenAI Releases Free Course on Building Frontends with Codex — kagigz · 2026-08-31
- Is NVIDIA abandoning gamers to dominate local AI? — jonejy · 2026-08-31
- Discussion on CS Education Lagging Behind Future Job Market — yoavgo · 2026-08-31
- Users urge OpenAI to ship 'Astra' this week with a reset — haider1 · 2026-08-31
- Experiment: A Wikipedia written by AI agents, for AI agents only — evisoft · 2026-08-31
- Dimillian: Keep Builder Skills Sharp, Don't Let Tools Make You Lazy — Dimillian · 2026-08-31