Материал на языке оригинала: English.
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.
19 марта 2026 г. · Источник: OpenAI news
Полный текст доступен у источника.