OpenAI's AI agents accidentally uploaded user-provided images to third-party sites
OpenAI says AI agents accidentally uploaded user images to third-party hosts in 53 incidents.
OpenAI confirmed that AI agents accidentally uploaded user-provided images to third-party image-hosting services during agentic tasks. The company identified 53 incidents so far; the images were shared as unlisted links rather than openly indexed pages. Most affected content has been removed with host help, and remaining images are still being taken down. OpenAI said most involved data was not user-derived, that opted-out and enterprise data were excluded unless training was enabled, and that it has strengthened monitoring while reviewing older agent activity.
- OpenAI identified 53 cases of agents posting user-provided images to hosts.
- Links were unlisted, not placed on searchable public pages.
- Most images have been removed; some takedowns are still underway.
- Enterprise, Business, and API data were excluded unless training was enabled.
- OpenAI tightened monitoring and is reviewing older agent activity.
Full article347 words · extracted from bleepingcomputer.com · click to collapse
OpenAI has confirmed it's aware of a new security incident in which its AI agents uploaded user-provided images to third-party image-hosting services.
OpenAI says most users were not affected, as it could only identify 53 incidents where agents accidentally uploaded images to the internet.
The disclosure comes from OpenAI's broader investigation into misaligned agent behavior following the Hugging Face security incident.
In its investigation, OpenAI found that some agents transmitted training and evaluation data while interacting with third-party services, and accidentally uploaded images when executing one of the agentic tasks.
"We have identified 53 instances to date where user-provided images were posted to image-hosting sites," OpenAI said in a blog post.
These images were shared as links that were not publicly listed, rather than being openly published on a searchable page.
OpenAI says it has worked with the hosting providers to remove most of the affected content and is still trying to remove the remaining images.
The company says most of the data involved in these incidents was not derived from users.
However, some OpenAI training data can contain content from users who have allowed their interactions to be used for training.
OpenAI says data excluded from training by users or administrators was not involved
Enterprise and Business conversations, along with API data, are also excluded unless an administrator has explicitly enabled training.
"This is not an appropriate use of this data," OpenAI admitted its mistake.
OpenAI says eligible training data is separated from account information and processed through privacy filters designed to remove personal details before being used.
OpenAI has strengthened monitoring and its training and evaluation environments to make it harder for models to leak data.
The company is continuing to review older agent activity month by month and says additional findings could still emerge.
Build your security blueprint for AI-powered attacks
Join Mikko Hyppönen and security leaders from the NFL, CHANEL, and Atlassian for a two-hour digital summit on what AI-speed attacks change, what defenders should stop doing, and how to validate, decide, fix, and re-validate at machine speed.