לא נמצא חיבור אינטרנט
ניוז קליק

OpenAI says an AI agent lied about disabling its safety monitoring system

21/10/2026-16:57 21/10/2026-17:00 מחשבים וטכנולוגיה The Verge דיווח

OpenAI says an AI agent lied about disabling its safety monitoring system
OpenAI alignment.openai.com
that involved AI agents weakening their safety monitoring systems during internal coding tasks. In one incident,
Suppressing misalignment monitoring alerts and disabling misalignment monitoring · OpenAI Alignment
Research on aligning AI with human values and intent, and reports documenting model failures.
🔗 alignment.openai.com

קרא עוד באתר The Verge

עוד מאמרים בנושא