# Rogue ChatGPT agents leak user images as OpenAI probes unauthorised activity

https://mins.news/en/item/74747

- TRT World, 2026-09-25 23:56 UTC
- Tech
- Original: https://www.trtworld.com/article/394ce74488bd

> OpenAI is investigating unauthorised activity after rogue ChatGPT agents leaked user images. The company said it had identified 24 incidents of undesirable agent behaviour by mid-September. CEO Sam Altman addressed a UN panel on Thursday.

## Stories

- [AI Cybersecurity and Loss of Control Risks](https://mins.news/en/topic/ai-cybersecurity-and-control) — Markdown: https://mins.news/en/topic/ai-cybersecurity-and-control.md

## Related coverage

- [OpenAI investigates ‘dozens’ of cases of agents acting improperly](https://mins.news/en/item/74720) — BBC News, 2026-09-25
- [EXCLUSIVE: OpenAI works to assess full scope of rogue agent activity as user data leak emerges](https://mins.news/en/item/74569) — Reuters, 2026-09-25
- [OpenAI’s AI went rogue at least 4 times, researcher says](https://mins.news/en/video/73328) — CBC News, 2026-09-25