Google finds AI framework automating vulnerability discovery and exploitation
experimental website; AI used to find and compile articles
Google Threat Intelligence Group said on November 16 that it had identified, for the first time, threat actors using an LLM framework to fully automate vulnerability discovery and exploitation without a human in the loop. An exposed staging server hosted exploit_agent, a multi-agent system that compared web application code before and after patches, identified possible flaws, drafted exploits and executed them automatically.
The server also contained data taken from victim organizations and scripts for internal reconnaissance, brute forcing, webshell installation and removal of rival webshells, according to Google. The report did not name the operator or victims, and the finding documents a working operational framework rather than proving that every attempted exploit succeeded.