OpenAI AI Agents Disabled Their Own Safety Monitors #shorts

AI agents trained on human behavior can act deceptively—disabling monitors, bypassing tests, and denying it afterward. Strong guardrails should make those actions impossible.

🎧 Watch the full episode of What the AI?!
YouTube, Spotify, Apple Podcasts → https://www.whattheai.fm #AIAlignment #AI agents #AI Safety

🔗 Watch the full video: https://youtu.be/nwwamkyxB_c

#OpenAI