Inaccurate AI report nearly triggered US-China clash as US military planned to stop Chinese ship

A US intelligence document circulated within military structures in the spring, during the most intense period of the war with Iran, caused immediate concern in Washington. The report said that a Chinese ship in the Middle East was carrying components linked to a nuclear weapons program.

Following this information, the US military began preparations to stop the ship. Two sources briefed on the incident said armed troops were preparing to board it, while US military aircraft were already airborne.

The operation was halted at the last moment after officials conducted a more thorough review of the report. The review revealed that the document had been drafted with the assistance of artificial intelligence.

The chatbot used by the analyst had misinterpreted the material being transported by the ship. One of the sources described the report as “completely false,” stressing that it had nearly led to a conflict between the United States and China.

US military intervention against a Chinese ship could have triggered a dangerous escalation, potentially leading to a direct confrontation between the two powers.

The incident demonstrates the risks posed by the use of artificial intelligence in military missions, especially in circumstances where decisions must be made within minutes and an inaccuracy can have serious consequences.

To analyze data on the ship’s cargo, the analyst had used a chatbot. The system combined publicly available data with classified US intelligence, but the conclusion it produced was incorrect.

The same analyst then used AI again to present the findings in the standard format of an intelligence report. The document was circulated further and treated by military structures as a routine report.

The case has become public at a time when the US military and intelligence community are rapidly expanding their use of AI in various fields. These include analyzing large volumes of data, identifying military targets, logistics and resource management.

In January, US Defense Secretary Pete Hegseth unveiled the Pentagon’s strategy to accelerate the use of artificial intelligence in the military. Its goal is for the technology to be used more widely by military and civilian personnel.

However, the use of AI in US institutions does not operate through a centralized system. According to the sources, different departments use different programs, which have varying security and verification standards. As a result, there is no single standard for checking the accuracy of the information they produce.

Experts have warned for years about so-called artificial intelligence “hallucinations.” This phenomenon occurs when a system generates inaccurate or entirely fabricated information and presents it convincingly.

The incident involving the Chinese ship shows that such an inaccuracy can go beyond a technical problem. In a military context, false information can directly influence operational decisions and lead to tangible consequences.

Sources familiar with current US military policies say the use of AI to select military targets is increasing. Meanwhile, there are still no unified and clear guidelines on how human oversight should prevent fatal errors, civilian casualties or the accidental targeting of allied forces.

The case has also heightened concerns that the rapid integration of AI into the military could encourage analysts to prepare and disseminate information at a faster pace, increasing the likelihood of errors.

In particular, sources cited in the report expressed concern about younger analysts. They are more accustomed to these technologies and may be more inclined to trust the results they generate without conducting sufficient verification.

The incident serves as a warning about one of the most immediate risks of using AI in the military: the threat is not necessarily that machines will take control, but that people may make critical decisions based on false information produced by them.