AI chatbot hallucination nearly triggered a US military strike on a Chinese ship
A CNN investigation found that a Pentagon analyst's AI chatbot invented a false report claiming a Chinese ship carried nuclear weapons materials, nearly triggering a US military strike before a routine review caught the error.
According to a CNN exclusive, a US Special Operations Command Pacific analyst used an AI chatbot in spring 2026, during the US-Iran war, to analyze the cargo manifest of a Chinese ship sailing in the Middle East. The chatbot combined open-source information with classified signals intelligence and falsely concluded the ship was carrying nuclear weapons program components. The analyst then used the same AI tool again to format the fabricated finding into a standard-looking military intelligence report, which circulated and set off alarm bells. According to sources, armed US troops prepared to board the ship and military aircraft were already airborne before a routine additional review caught the error and the operation was called off. One source described the report as entirely false and said it almost started a war with China. The specific chatbot involved has not been publicly identified. The incident comes as Defense Secretary Pete Hegseth pushes for wider adoption of AI tools across the Pentagon.