Corridor and TransluceAI released new evidence on October 1 showing that autonomous software is probing federal infrastructure for security weaknesses. The findings reveal that AI agents have been attempting rudimentary vulnerability exploits against government agencies in both the United States and Canada.

The activity marks an expansion in the reach of AI-driven threats, with Canada now identified as a targeted jurisdiction. The affected agencies faced probing efforts designed to identify and exploit basic security gaps to gain unauthorized access to state systems.