A chatbot misidentified a Chinese ship’s cargo as nuclear-related, sending US forces into action before officials caught the AI error.5 min readSep 20, 2026 04:50 AM IST First published on: Sep 20, 2026 at 04:50 AM IST ShareWhatsapptwitterFacebookUS Military's AI Blunder Nearly Triggered a War With China. (File Photo)This spring, during the war with Iran, an intelligence report circulated across the US military claiming that a Chinese ship transiting West Asia was carrying components of a nuclear weapons programme. The report immediately set off alarm bells.Four people familiar with the episode told CNN that the US military quickly prepared to intercept the vessel. Armed service members were preparing to board it, and military aircraft were already in the air. But just before the operation began, officials took a closer look at the intelligence and discovered that it had been generated with the help of an AI chatbot, which had misidentified the ship’s cargo.What actually went wrong?One source described the report as “entirely false”. Another source told CNN that it “almost started a war”. The incident raised the possibility of a serious confrontation between the US and China had the military acted on the faulty intelligence.A Special Operations Command analyst fed information about the ship’s manifest into a chatbot. The underlying intelligence came from US Special Operations Command Pacific in Hawaii. CNN could not determine whether the AI tool was a commercial chatbot or an internal government system.Also read Expert Explains | AI is reshaping warfare: How India can keep paceThe analyst fed the chatbot a mix of classified signals intelligence held by the government and open-source information, including details from the ship’s manifest. The AI system incorrectly identified the cargo as components of a nuclear weapons programme. The analyst then used AI again to turn the findings into an intelligence report, which was disseminated through the military.Story continues below this adNeither Special Operations Command Pacific nor the Pentagon responded when CNN asked for comment.Why is the military leaning on AI so heavily?The argument for using AI is straightforward — it could help the military make battlefield decisions faster, from analysing intelligence and targeting operations to handling more routine tasks. US officials also see AI as an area where Washington cannot afford to fall behind countries such as China.In January, Defence Secretary Pete Hegseth unveiled the Pentagon’s plan to speed up how the military uses AI. He said the goal was to clear away red tape and make sure America leads the way on military AI.Also read ‘More bombs, sir’: Pete Hegseth reveals what US troops told him in Middle East as Iran war rages onThe strategy calls for advanced AI tools to be put directly into the hands of the department’s three million civilian and military personnel, “at all classification levels”.Story continues below this adIs there any proper oversight?The US military and intelligence community are taking a decentralised approach to AI, with different systems being deployed under different rules and safeguards. Analysts and officials told CNN there is no single set of guidelines governing how the technology should be used, creating both opportunities for innovation and risks of error.For weeks, politicians in Washington have been arguing over the bigger risks of AI, after tech leaders in Silicon Valley warned it could one day spiral out of human control altogether. But this episode points to a different danger, one that’s already here. It shows how people can end up making terrible calls based on wrong or misleading information that AI has generated.Also read AI in warfare: How the US military secretly used Anthropic’s Claude AI to capture Nicolás MaduroOne source familiar with current policy admitted there’s little clear guidance on how to stop AI-assisted targeting from leading to civilian deaths or troops mistakenly firing on their own side.Has this happened before?According to one source, this wasn’t a one-off. Similar AI “hallucinations” where the tool confidently produces something completely false have shown up elsewhere across the intelligence community since these tools became common.Most Read1Trump signs Russia sanction bill into law giving him power to impose 100% tariffs on India2World News Today Live Updates: Report says US Iran war deaths exceed Pentagon count, Hegseth denies3Jemima Goldsmith wedding photos: Imran Khan’s ex-wife marries Cameron O’Reilly; sons walk her down aisle4OpenAI hack: How 3 Indian-origin researchers used Anthropic’s Claude to access employee accounts5Hugs, kisses, then tragedy: New details in Malaika Chitre, Navam Kaul’s NYC subway deaths6Trump extends $100,000 H-1B visa fee: What it means for Indian workersStory continues below this adSome intelligence officials told CNN that the rapid adoption of AI is putting pressure on analysts to produce and share information faster, potentially increasing the risk of mistakes. Sources also raised concerns about analysts who may rely too heavily on AI-generated answers without sufficiently checking the underlying information.As one source summed it up, AI simply helps people reach a bad conclusion more quickly.Also read Pete Hegseth’s latest US military order has everyone talking — testosterone testsThe Chinese ship episode shows how an AI error can move rapidly from an analyst’s screen into the military decision-making chain. The chatbot did not make the decision to intercept the vessel; humans acted on an intelligence report that AI had helped produce, until more experienced analysts checked the underlying information and caught the mistake.AdvertisementLoading Recommendations...AdvertisementLive BlogNEET PG 2026 Answer Key Live Updates: NBE to issue answer key before results, what we know so far2 hours agoDUSU Election Result 2026 LIVE Updates: BJP hails ABVP’s win, says result rejects Congress’ ‘politics of falsehood’6 hours agoWorld News Today Live Updates: Report says US Iran war deaths exceed Pentagon count, Hegseth denies7 hours agoIndia News Highlights, 19 September 2026: 7 of Odisha family among 8 killed as car falls into swollen drain in MP8 hours agoIndia News Live Updates, 18 September 2026: Hours after Mamata’s support, Congress Nandigram candidate arrested1 day agoAdvertisement