Google and Mandiant find attackers using autonomous AI agent framework to turn security patches into exploits
Google Threat Intelligence Group and Mandiant reported the first observed case of attackers using an autonomous AI framework to turn newly released security patches into working exploits with no human in the loop. Researchers found a publicly exposed staging server running a multi-agent framework called exploit_agent that connected frontier and locally hosted models to automate patch diffing, exploit generation and execution. The server also held stolen victim data and post-compromise scripts for reconnaissance, brute-force attacks and webshells.