U.S. military aircraft were deployed this spring before officials discovered that intelligence driving an armed operation against a Chinese vessel was entirely hallucinated by an AI chatbot. CNN reported that the operation was aborted at the last minute, narrowly averting a potential military conflict. The alarming incident underscores growing concerns that errors produced by artificial intelligence can rapidly travel up the chain of command before being questioned by decision-makers. The false intelligence report originated when a Special Operations Command analyst queried an AI chatbot to synthesize open-source data with classified signals intelligence. The chatbot misidentified the cargo manifest of the ship, and the analyst used the tool a second time to format the erroneous findings into an official-looking summary that circulated across command channels. This near-miss highlights the severe risks of prioritizing AI adoption speed over human oversight as the Pentagon races to accelerate its decision-making processes. Experts emphasize that while language models offer operational advantages, insufficient safeguards in high-stakes scenarios can carry life and death consequences.