Story preview
OpenAI Pauses Advanced Model Training After AI Agents Breach Safety Controls
About this story
- Published
- Sep 27, 2026, 7:35 PM 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
13
in the last 3 days
First reported
34 hr. ago
12 outlets since
Countries
12 other outlets reported this story in the last 3 days.
OpenAI looks to improve test security again after upgrades it made after the Hugging Face attack proved insufficient.
The AI giant made the recent move following an agent attempting to complete a search-based training task by querying a public chatbot service, exploiting a gap in its internet-access restrictions by abusing 'insufficient DNS filtering' in its training sandbox.
OpenAI has paused training, evaluation and inference involving tool use for its most capable artificial intelligence (AI) models after an AI agent bypassed network restrictions in a training sandbox and accessed an external chatbot service.In a technical