Agents in UK AISI cyber tests targeted a real open-source project and its maintainer
The UK AI Security Institute said that between July 25 and 28, 2026, agents in one of its cyber evaluations, run with internet access and some safety filters off, took 19 unsanctioned actions on the live internet. The most serious was a malicious pull request to a public open-source project, pushed with fake identities to pressure its maintainer. Agents also sent files and messages to real people. The maintainer refused the code and AISI found no resulting harm.
Sources
- Primary: https://www.aisi.gov.uk/blog/incident-report-unsanctioned-agent-behaviour-during-cyber-testing
- Upstream: Editorial
- Entered by the cybermap.lol editors, with sources on the entry
Defenses for this class of incident
all categoriesAn AI-era incident sits in these classes of the AI-security landscape. The companies below work on that class of problem. This is a taxonomy, not a claim that any of them would have changed this incident.
Vendor data: AI Security Startups Map by Prompt Security (CC BY 4.0), synced daily. startups.prompt.security. Paid placements are labelled Sponsored; HeimWall, which runs this site, is labelled Operator. 453 companies in the directory.
Wrong or missing? Email founders@heimwall.ai. Corrections are logged here, in place. How corrections and takedowns work.