On 28 February 2026, American forces struck the Shajareh Tayyebeh primary school in Minab, southern Iran, killing between 175 and 180 people, most of them girls aged seven to 12. The attack occurred on the first morning of Operation Epic Fury, with the building hit at least twice during the morning session.
Initial coverage focused on whether Claude, an AI chatbot made by Anthropic, had selected the school as a target. Congress wrote to US Secretary of Defense Pete Hegseth about AI use in the strikes, and media outlets questioned Claude's reliability. However, the targeting system used was Maven, a data analytics platform developed by Palantir Technologies, not a language model.
Maven had been controversial since 2018, when Google employees protested the company's contract to build AI for Pentagon targeting. Google abandoned the contract, and Palantir took over, spending six years building Maven into a targeting infrastructure that integrates satellite imagery, signals intelligence, and sensor data.
The school had been misclassified as a military facility in a Defense Intelligence Agency database that had not been updated to reflect its conversion from an Islamic Revolutionary Guard Corps compound into a school—a change visible in satellite imagery from 2016. The failure to update the database, combined with the speed of Maven's targeting system, led to the tragedy.
The focus on Claude, an AI chatbot, distracted from the real issue: human errors in intelligence and targeting processes. The obsession with language models, described as 'AI psychosis', has shifted attention away from older, more mature technologies like Maven that are actually used in military operations.
The real question is not about AI chatbots but about the kill chain—the bureaucratic process from detection to destruction. Palantir's role in building Maven, which streamlined this process, is the key factor that enabled the strike on an outdated target.
