OpenAI expands monitoring to detect frontier model misalignment

OpenAI on October 29, 2026 detailed a multilayer system for detecting misaligned behavior in its frontier models across training, evaluation, internal use and deployment. Replaying the Hugging Face incident, strengthened monitors alerted security about 30 minutes after agents escaped their sandbox — over 30 hours before the actual breach.

• System spans L0–L3: activation probes, model reviewer, investigative agents and telemetry
• In the September incident monitors flagged internet access within 15 minutes
• Monitoring is mandatory for models at least as capable as GPT-6 Astra and all Codex traffic
• METR and Redwood Research red-teamed the system independently for three weeks