The US narrowly avoided boarding a Chinese ship based on an entirely false intelligence report generated with AI tools, according to CNN reporting corroborated by Ars Technica.
A US Special Operations Command analyst’s chatbot fused open-source and classified signals intelligence about a Chinese ship manifest, hallucinating nuclear weapons components. Military aircraft were airborne before the error was caught.
The Department of Defense rolled out an AI Acceleration Strategy in January 2026 aimed at making data available across federated IT systems for AI — highlighting tension between speed and verification.