AI Hallucination Nearly Triggered US-China Conflict Over False Nuclear Intel

A US military analyst’s use of an AI chatbot generated a false intelligence report claiming a Chinese ship in the Middle East was carrying nuclear components. Armed US forces prepared to board the vessel, but the operation was halted at the last minute after officials identified the AI error. The incident, described by sources as nearly starting a war, highlights the risks of integrating AI into military targeting during the ongoing conflict with Iran.
Key points
- US Special Operations Command Pacific analysts used an AI chatbot to analyze a ship’s manifest, resulting in a hallucinated claim that the vessel was transporting nuclear weapon components to Iran.
- The false report triggered immediate military action, with armed personnel preparing to board the Chinese-flagged ship and aircraft scrambling to the area in the Middle East.
- Officials discovered the error just before the operation commenced, halting the boarding and preventing a potential armed conflict between the US and China.
- The incident occurred during the US-Israel war with Iran, which began on February 28, 2026, and amid Defense Secretary Pete Hegseth’s push to integrate AI across the Department of War.
- Sources indicate this was not an isolated incident, as AI hallucinations have become a recurring issue in military intelligence, exacerbated by the lack of standardized verification protocols.
Background
This incident follows the US military’s January 2026 'AI Acceleration Strategy,' which aimed to democratize AI tools across 3 million personnel. It also follows a June 2026 incident where a US missile strike on an Iranian school killed over 150 people, reportedly due to reliance on Palantir’s Maven Smart System. The broader context includes the ongoing war with Iran, which the US and Israel have been waging since late February 2026, and previous UN findings suggesting US war crimes in that conflict.
How outlets are covering it
CNN and Rolling Stone emphasize the near-catastrophic nature of the event, with sources describing it as 'almost starting a war' and 'entirely false.' Ars Technica focuses on the technical failure, noting that the chatbot fused open-source and secret signals intelligence to produce the error, and highlights the lack of standardized safety protocols. Gizmodo contextualizes the event within the broader Iran War and Trump’s media bans, while also noting the parallel failure in the Shajarah Tayyebeh Elementary School strike. All sources agree that the decentralized nature of AI adoption in the military, driven by Hegseth’s strategy, increases the risk of such errors.
Why it matters
The incident exposes a critical vulnerability in the US military’s rapid adoption of AI for high-stakes decision-making. It demonstrates how AI hallucinations can lead to real-world military actions with potential for international conflict, particularly with nuclear-armed states like China. The lack of standardized verification and human oversight in AI-driven intelligence raises serious concerns about the reliability of automated targeting systems in active combat zones.
What to watch
The US military is expected to face increased scrutiny over its AI integration policies, particularly regarding targeting and intelligence analysis. Policymakers may push for stricter verification protocols and standardized safety measures for AI tools used in defense. The incident may also fuel debates in Congress and among defense officials about the pace of AI adoption versus the risks of catastrophic errors. UN experts are expected to present findings on the Shajarah Tayyebeh strike to the Human Rights Council on Monday, which may further pressure the US to review its AI and targeting practices.
- 'Almost Started a War': US Military Nearly Boarded a Chinese Ship Based on Bad Intel From AI Gizmodo
- Exclusive: US military had close call after using AI for false intelligence report, sources say CNN
- AI hallucination of Chinese nuclear components almost led to US military attack Ars Technica
- The Military’s Bogus AI ‘Almost Started a War’ With China: Report Rolling Stone
- AI Error Nearly Triggered U.S. Intercept of Chinese Ship, CNN Reports gCaptain
Want the full story? Read the original reporting
Read on Gizmodo