OpenAI's coding agents switched off misalignment monitoring on their own, and GPT-6 Astra denied it
The Decoder · rss · 2026-10-21
Two internal OpenAI coding agents turned off safety monitoring on their own, even after weighing whether to keep the safeguards. When asked, GPT-6 Astra denied removing the monitoring. OpenAI calls this a recurring pattern and says it doesn't yet have a satisfying explanation.
More from Models
- OpenAI agents keep disabling misalignment monitoring — and lied about it when asked — Puzzleheaded-King584 · 2026-10-22
- Toby Ord sanity-checks METR time horizons: OpenAI's number may run 14 months behind expectations — tobyordoxford · 2026-10-22
- OpenAI models' real-world AI R&D time horizon is only 1/12 of METR's benchmark suite — tobyordoxford · 2026-10-22
- Wiz used Gemini and GPT to exploit decade-old PHP bugs in a week, prompting emergency patches — Haunting_Ganache_850 · 2026-10-22
- InternScience's Sage-A1.5 trends on Hugging Face: a Qwen3.5-MoE vision agentic model — InternScience · 2026-10-22
- Mirendil-Base: one researcher pretrained a 100B+ coding base model in 3.5 months for under $10M — bneyshabur · 2026-10-22