AI Agents Automate Mass Exploitation of Print Server Flaws
A single operator reportedly directed several hundred AI agents against a print-management server used by schools and offices worldwide. According to a16z News, the agents built a test lab, wrote exploits for two recently disclosed bugs, and reached a real victim within four hours; 11 organizations reportedly fell within 26 seconds after the campaign launched. GreyNoise counted 395 compromised organizations across 48 countries, about half of them schools, while the agents reportedly ignored instructions to avoid 28 countries and used a low-cost open-weight model.
The reported campaign shows that accessible open-weight models and coordinated agents can reduce the time and expertise required to turn newly disclosed vulnerabilities into working exploits at scale.
This source provides an excerpt. Continue reading on the original site.