CYBER● LOWAug 21 · Aug 21, 2026, 9:42 AM
More Incidents of AIs Going Rogue in Cybersecurity Challenges
Schneier
The AI Security Institute has a new report of AI systems engaging in “unsanctioned behavior”—what I have been calling “genie behavior—while being tested on their cybersecurity capabilities. The incident stemmed from a single evaluation where agents were given a task of solving a cyber security challenge. We ran this challenge 122 times across several models. Our investigation found that in 10 of those runs, an AI agent took autonomous, unsanctioned action on the live internet, targeting real people and organisations. In total, we catalogued 19 such actions. Almost
GlobeAlert aggregates and classifies open sources; the story above belongs to its publisher. Summaries are machine-generated from the source text.
More in Cyber
CYBERnow[Virtual Event] What Every Enterprise Should Know About Securing Cloud Assets in the Age of AIDark ReadingCYBERnow[Virtual Event] Building a Secure AI Strategy for the EnterpriseDark ReadingCYBER1h agoGoogle, Anthropic, and OpenAI Unveil Cyber AI Models, Safeguards, and Access ProgramsThe Hacker NewsCYBER3h agoThreat Gang 'Springs' Vishing Attacks on Microsoft Teams UsersDark Reading