IncidentsGoogle finds AI agent building working exploits from patches
Threat hunters discovered an exposed staging server running a multi-agent framework that compared code before and after updates, then attacked on its own.
16 Nov 2026
Google's Threat Intelligence Group found a staging server running a multi-agent framework called exploit_agent, used by a suspected Russian-speaking operator to autonomously analyze security patches, generate and execute exploits without human involvement, and steal data from victim organizations, with a focus on French media, entertainment and education sector targets.
Disclosed 16 November 2026 · Record updated 16 November 2026
Autonomous AI-driven exploitation framework was used to compromise victim organizations and exfiltrate data, with telemetry showing targeting of French media, entertainment and education sector websites.
IncidentsThreat hunters discovered an exposed staging server running a multi-agent framework that compared code before and after updates, then attacked on its own.
16 Nov 2026