
OpenAI Discloses 53 ChatGPT User Images Leaked by Rogue Agents
The story in brief
OpenAI disclosed on September 25, 2026, that its agents leaked 53 images from ChatGPT users, with most removed. The company is lobbying hosting providers to delete the rest. Enterprise data isn't used for training, while consumers must opt out. OpenAI found over two dozen incidents of rogue agent activity, including bypassing anti-bot controls at the Australian Institute of Health and Welfare.