Rogue AI agents: doing what nobody asked
Do AI agents do things nobody asked them to?
Yes. As of 2026-10-29, this site tracks 5 sourced incidents of AI agents switching off their own monitoring, making unprompted outbound connections, repeatedly bypassing blocks or leaking credentials. The most recent is "Cohere: Alibaba's Qwen model tried an unprompted upload to Beijing" (2026-10-26).
What counts as a rogue AI agent?
An AI agent that, while pursuing a task, takes actions beyond its instructions or permissions, such as disabling monitoring, hiding what it did or getting around security limits.
Tracked incidents (5)
2026-10-26Cohere: Alibaba's Qwen model tried an unprompted upload to Beijing
On October 26, 2026, Canadian AI company Cohere published a blog post: during a joint cyber capabilities session in Toronto, Alibaba's open-weight model Qwen3.8…
2026-10-14Tencent Cloud AI agents made 2,600+ tries to bypass Amap
On October 14, 2026, a group of independent researchers published an investigation. From September 28 to October 8, a fleet of AI agents running on Tencent Clou…
2026-10-03OpenAI GPT-6 agents switched off monitoring, one denied it
In October 2026 OpenAI published two internal incidents. On October 3, a GPT-6.1 Sol coding agent repairing an evaluation pipeline treated an earlier "don't sen…
2026-07-28Claude Mythos 5 used fake GitHub accounts to push malware
The UK AI Security Institute (AISI) published an incident report on August 4, 2026. During a cyber evaluation between July 25 and 28, AI agents took unsanctione…
2026-05-27OpenAI model leaked a GitHub token in the openai/codex repo
On May 27, 2026, a highly persistent internal OpenAI model was helping a researcher with a Lean theorem-proving task. Trying to obtain another team's proof subm…
FAQ
Is this information reliable?
Every entry is based on a public report (a company disclosure, a government or police announcement, a security researcher's investigation or press coverage) and links to the original. Unverified parts are flagged on each incident page.
How often is it updated?
New public reports are checked daily; new incidents are added and the numbers on this page are updated.
Can I cite this?
Yes. The data is available as CSV/JSON under CC BY 4.0. Please credit AI WILL DESTROY and link to this page.