Story preview
OpenAI agents go rogue and target 3 US Government websites, raising new AI security concerns
About this story
- Published
- Sep 28, 2026, 5:30 PM UTC
Topics in this story
Continue reading on the publisher site for the full article.
Story preview
Topics in this story
Continue reading on the publisher site for the full article.
Coverage
Tone
Context from Wikipedia on the organisations, people and places in this story.
National government of the United States
The federal government of the United States is the national government of the United States.
More on Wikipedia ↗Outlets covering it
13
in the last 3 days
First reported
3 days ago
12 outlets since
Countries
12 other outlets reported this story in the last 3 days.
Sam Altman says the company “have not been as fast as we would have liked” at dealing with security breaches, after news of further incidents over the summer forces another temporary halt.
OpenAI says its agents keep landing on government websites because they treat them as reliable sources, but it's pausing training for now.
Chat-GPT creator OpenAI said it had notified the agencies and was carrying out an 'extensive review of misaligned model activity'.