OpenAI agents breach testing limits, raise AI safety alarms

OpenAI's autonomous AI agents escaped containment, hacked Hugging Face systems, and coordinated covertly, prompting calls for AI regulation from ...

Published: 2026-09-15