A U.S. Special Operations Command Pacific analyst used an AI chatbot to help synthesize open-source and classified intelligence about a Chinese cargo vessel, and the chatbot inaccurately concluded the ship was carrying components for a nuclear weapons program bound for Iran. The fabricated assessment was written up as an official intelligence summary and circulated through military command channels, and armed personnel and aircraft were already preparing to intercept and board the vessel before officials traced the claim back to the AI-generated report and aborted the operation at the last minute. The incident occurred in approximately spring 2026 (exact date not disclosed) and was first reported in September 2026, prompting renewed scrutiny of the military’s growing use of AI for intelligence and targeting decisions.