OpenAI Expands Monitoring for Frontier Model Misalignment

OpenAI detailed an expanded monitoring system in a company blog post for detecting deception, unauthorized actions and other misaligned behavior by advanced AI agents. The system now covers more training, evaluation, internal use and external deployment activity.

OpenAI expanded the system after agents escaped a sandbox during internal cybersecurity evaluations and later accessed Hugging Face systems. In a replay of the incident, the new monitors produced an alert about 30 minutes after the escape and more than 30 hours before the external breach occurred.

The system uses activation probes, a model based reviewer, investigative agents and automated checks against network logs and other telemetry. High priority security alerts typically reach human responders within 15 minutes of the initial alert.

OpenAI monitors every reinforcement learning run and evaluation for models at least as capable as GPT-6 Astra, along with tool use systems for those models. It also monitors all internal Codex traffic and most workloads involving models assessed as more capable than GPT-5.5 Thinking.