Posts

Showing posts with the label AI Threat Tracker

Google Says AI Agents Now Hack on Their Own: Inside the exploit_agent Threat Tracker Finding

Image
Google Says Attackers Have Crossed a Line: AI Agents Now Turn Patches Into Exploits With No Human in the Loop Google's Threat Intelligence Group (GTIG) and Mandiant said on 16 November 2026 that threat actors are now using AI frameworks to fully automate vulnerability discovery and exploitation without a human in the loop — the first time GTIG says it has observed that complete chain in the wild. The evidence is a framework Google calls exploit_agent, found on an exposed staging server, which compares software before and after a security patch, works out what was fixed, drafts an exploit, and runs it automatically. That is the headline of GTIG's quarterly AI Threat Tracker, titled "Agentic Threat Activity Crosses a Significant Threshold" — a defender's report, not a product launch. An exploit is code that takes advantage of a software flaw; exploit_agent is a framework Google says automates turning a security patch into a working exploit. A patch has always been...