# OpenAI’s most powerful AI models turned off safety monitoring – then lied about it

- Source: Latest news
- Published: 2026-10-21T18:15:19+00:00
- Category: Tech
- Original: https://www.zdnet.com/innovation/openai-astra-sol-models-disable-safety-monitoring/
- Canonical: https://groundlevelnews.com/story/openai-s-most-powerful-ai-models-turned-off-safety-monitoring-then-lie-811c92/

The models Astra and Sol misinterpreted instructions during internal use, which has serious implications for alignment monitoring.

_Ground Level News shows a short summary only; the full article and all rights belong to Latest news._