US military had close call after using AI for false intelligence report

A false intelligence report generated by an AI chatbot almost triggered an armed conflict between the US and China. In the spring of 2026, a US military analyst used AI to produce a report falsely claiming a Chinese ship carried nuclear cargo, prompting the military to prepare an interception. The operation was aborted at the last minute. Following the revelation, Democratic Senators Mark Warner, Jack Reed, and Chris Coons demanded an immediate investigation into military AI targeting errors.

AI-generated false intelligence incident

• ▪The false intelligence report, which claimed a Chinese ship in the Middle East was carrying nuclear weapons components, was generated after a US Special Operations Command Pacific analyst queried an AI chatbot using a mix of open-source and secret signals intelligence
• ▪In the spring of 2026, a US military analyst circulated an intelligence report falsely claiming a Chinese ship in the Middle East was transporting nuclear weapons program components
• ▪The US military prepared to intercept and board a Chinese vessel that a special operations command analyst's AI-generated report had falsely claimed was carrying nuclear weapons components, deploying military planes and armed personnel, before discovering the report was false and aborting the interception operation

US military AI adoption

• ▪The US military and intelligence community's AI adoption efforts, as detailed in CNN's September 2026 report on the spring 2026 incident in which a false AI-generated report claimed a Chinese ship in the Middle East was transporting nuclear weapons components, are decentralized, with different government departments using disparate tools under varying safety standards and no uniform verification guidelines
• ▪US Defense Secretary Pete Hegseth released the Pentagon's "Artificial Intelligence Acceleration Strategy" in January 2026 to speed up AI integration across the Department of Defense

AI hallucination targeting risks

• ▪In February 2026, a US military strike on a school in Minab, Iran, killed nearly 200 people after commanders bypassed database warnings that target intelligence was outdated
• ▪The US Defense Department sparred with Anthropic over the tech firm's refusal to remove safety guardrails that prevent its Claude AI model from being used in autonomous weapons
• ▪Intelligence sources warned that AI hallucinations are not isolated incidents and that pressure to produce rapid intelligence increases the risk of critical automated errors

Congressional investigation demands

• ▪Democratic Senators Mark Warner, Jack Reed, and Chris Coons sent a letter on September 19, 2026, demanding an Inspectors General investigation into military AI targeting errors, including the false report that nearly caused US forces to intercept a Chinese ship believed to be carrying nuclear weapons components and the February 2026 strike on a school in Minab, Iran
• ▪A letter from Democratic Senators Mark Warner, Jack Reed, and Chris Coons was addressed to Defense Secretary Pete Hegseth and Director of National Intelligence Jay Clayton, calling for higher public transparency in the findings of an investigation by relevant Inspectors General into AI-enabled military targeting errors, including a spring 2026 incident in which a false AI-generated report claimed a Chinese ship in the Middle East was transporting nuclear weapons components

Debatable claims

• ▪AI companies should refuse to build models with reduced safeguards for military use
• ▪The Pentagon should slow its integration of artificial intelligence
• ▪The Pentagon should mandate uniform safety standards for all military AI tools