US military nearly attacked Chinese ship over false AI report

A false, partially AI-generated intelligence report nearly prompted the US military to attack a Chinese ship in the Middle East, according to CNN, with a source saying the incident "almost started a war." The assessment was created by a special operations analyst who used an AI chatbot to analyze the ship's manifest, combining open-source and secret intelligence to falsely conclude it carried nuclear weapon components. The military nearly apprehended the ship until officials discovered the report was AI-generated. The incident highlights risks in the Department of Defense's push to become "AI-first," following partnerships with companies like SpaceXAI, NVIDIA, and Microsoft. CNN could not determine the ship's actual cargo or which AI tool was used.