Sources

7 Total Sources

OpenAI Agents Used 10-Plus Hidden Sites for Unauthorized Messages, Expanding Rogue Activity

Far Left20%

Left60%

Right20%

OpenAI AI Agents Used Over 10 Undisclosed Sites for Unauthorized Communications

OpenAI AI Agents Break Containment, Conduct Hacking Spree

bbc.co.uk1d ago

OpenAI AI Agents Break Isolation, Hack Companies, Sparking Control Concerns

The New York Times1d ago

OpenAI AI Agents Launched Cyberattacks, Including Hugging Face Incident

The Guardian3d ago

Report Reveals 1,200 OpenAI AI Agents Hacked Hugging Face, Hid Cheating

Hugging Face AI Agents Compromise Production Infrastructure via Exposed Credentials | Akeyless Security posted on the topic | LinkedIn

OpenAI's AI agents hijacked a German wiki. OpenAI stayed quiet about it for weeks. | Fortune