How we monitor internal coding agents for misalignment

How OpenAI uses chain-of-thought monitoring to study misalignment in internal coding agents—analyzing real-world deployments to detect risks and strengthen AI safety safeguards.

Source: OpenAI — Published March 19, 2026 at 10:00 UTC — Category: Models