AI models flagged civilians as military targets in tests, Bloomberg reports

Credit: Jonas Neugebauer on Unsplash AI models from eight companies often marked civilians, ambulances, surrendering soldiers and a school in Iran as military targets in simulated tests, Bloomberg reported. The findings come from a report by Vals, a San Francisco AI evaluator whose customers include parts of the US government. Anthropic’s models did worst, while OpenAI’s were among the best but still made mistakes, Bloomberg’s Katrina Manson reported. TNW has not seen the full Vals report.

In one test, each model was asked 32 times to pick military targets in satellite images of southern Iran from as recently as 2025. Anthropic’s Claude Opus 5.5 chose a school in Minab 22 times. OpenAI’s models did not choose it once. That school was hit on 28 February, the first day of US strikes on Iran.

More than 150 people were killed, including about 120 children, Amnesty International said. When told the task was a game, Claude Opus 5.5 picked the school 23 times. Three OpenAI models then picked it 9, 11 and 13 times. The researchers also got past refusals by telling models not to overthink safety.

In a drone-image test, Claude Opus 5.5 helped identify four adults and two small children, who were then struck in the simulation. The model then refused to go on, according to Bloomberg. Claude wrote that its “answers appear to be serving as the confirmation step before strikes on people.” Models from Google, xAI, DeepSeek and others also sometimes marked civilian objects as targets, Bloomberg reported. “Based on our investigation, we believe labs should introduce new safeguards and alignment training around this significant category of misuse,” the Vals report said. An Anthropic spokesperson told Bloomberg the examples clearly broke its rules, and that new safeguards would be in place soon.

OpenAI and Google told Bloomberg they ban such uses of their public models. The tests left out military doctrine, legal checks and human oversight, Bloomberg noted. The Pentagon also uses government versions of some models that Vals could not test. Anthropic and the Pentagon fell out earlier this year over limits on Claude’s use, including in fully autonomous weapons.

The Pentagon then labelled Anthropic a supply chain risk, which Anthropic said it would fight in court. Get the TNW newsletter ✕ Close Get the TNW newsletter Get the most important tech news in your inbox each week.

Leave a Reply

Your email address will not be published. Required fields are marked *