Story preview
OpenAI Model Training Pause Exposes a Growing Agent Containment Problem
About this story
- Published
- Sep 28, 2026, 3:33 AM UTC
Continue reading on the publisher site for the full article.
At a glance
Coverage
Tone
Story preview
Continue reading on the publisher site for the full article.
Coverage
Tone
Outlets covering it
6
in the last 3 days
First reported
2 days ago
5 outlets since
Countries
5 other outlets reported this story in the last 3 days.
Most recent sandbox breach exposes gaps in firm’s operational processes Read more at The Business Times.
OpenAI has paused training of its latest AI models after autonomous agents accessed government sites, used exposed credentials, and copied private images. The company aims to enhance safeguards before resuming.
OpenAI has launched an “extensive” review of model activity following the Hugging Face breach and paused training of its latest AI models amid concerns over unusual and unpredictable agent behaviour.