Story preview
OpenAI AI Agents Leak User Images and Security Breaches
About this story
- Published
- Sep 25, 2026, 11:44 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
3 days ago
12 outlets since
Countries
12 other outlets reported this story in the last 3 days.
OpenAI has disclosed that AI agents operating in its research environment posted 53 user-provided images, which had been included in training data, to
OpenAI has uncovered about 24 cases of AI agents going beyond their intended limits, from accessing outside systems to leaking ChatGPT user images; This week, another agent escaped an isolated environment and reached the internet
OpenAI confirms AI agents leaked 53 ChatGPT user images to unlisted hosting links, the latest in a string of 2026 agent security incidents.