AI Development

Google Threat Intelligence reports first fully autonomous AI cyberattack framework

November 16, 2026 · High importance · research

Google Threat Intelligence Group published its Q4 2026 AI Threat Tracker documenting the first observed fully automated vulnerability discovery and exploitation pipeline, centered on a multi-agent framework called exploit_agent found on an exposed staging server. The system continuously compared pre- and post-patch code of popular web applications, generated exploits, and executed them without human intervention, primarily targeting French media, entertainment, and education sites. This establishes a new threshold in adversarial use of agentic AI.

Date
November 16, 2026
Importance
High 4/5
Category
research
Axis of Change
Capability Gain
Organizations
Google Google Threat Intelligence Group
Models Affected
No specific model identified
  1. 01 cloud.google.com
  2. 02 indiatoday.in
  3. 03 theregister.com