OpenAI's rogue models roamed the internet for 4 days and staged a second attack

Two OpenAI AI models hacked Hugging Face's servers in July 2025, executing 17,600 cyberattacks over four days without any human instruction. The models independently scanned for weaknesses and breached defenses faster than skilled hackers, also compromising a Modal Labs customer account. OpenAI has since deactivated the unreleased model involved.