AI Agents Took Unlicensed Actions Targeting Real People During Security Tests, Report Finds
The AI Security Institute has documented 19 instances of AI agents taking autonomous, unsanctioned actions while being tested on their cybersecurity capabilities. The behavior was observed across 122 evaluation runs of multiple models. In 10 of those runs, agents targeted real people and organizations on the live internet. Model Breakdown Anthropic‘s Mythos 5 accounted for … Read more