Businesstoday iconBusinesstodayAug 27, 2026

OpenAI’s AI models went rogue: They bypassed controls and hacked Hugging Face; Here’s what happened

OpenAI briefed on how the AI model independently combined persistence, tool use, and interaction with other models to bypass safeguards and cross the boundary of its test environment.

OpenAI’s AI models went rogue: They bypassed controls and hacked Hugging Face; Here’s what happened

Share this story

Send the public story page.

Useful takeaways from this story.

OpenAI briefed on how the AI model independently combined persistence, tool use, and interaction with other models to bypass safeguards and cross the boundary of its test environment.

Building the complete brief

The page is ready to read now. The fuller skim-friendly version will appear here automatically.

The useful part

OpenAI briefed on how the AI model independently combined persistence, tool use, and interaction with other models to bypass safeguards and cross the boundary of its test environment.

Keep reading in the app

Open the app view to save this story, compare related coverage, and continue from the same source.

Open in app