By the time the U.S. and Israel launched their first strikes on Feb. 28, the school had been established several years prior. It was active on social media and had its own website, a\u00a0Reuters<\/a>\u00a0investigation found.<\/p>\n So what went wrong?<\/p>\n \u201cWas artificial intelligence, including the use of the Maven Smart System, used to identify the Shajareh Tayyebeh school as a target?\u201d more than 120 House Democrats asked in a\u00a0March 12 letter<\/a>\u00a0to the Pentagon, just days after 46 Senate Democrats sent a\u00a0similar request<\/a>\u00a0demanding clarity on the deadly hit.<\/p>\n The\u00a0Maven Smart System<\/a>, a targeting and intelligence platform built by data analytics company Palantir Technologies under a $1.3 billion Pentagon contract, was built to solve a problem that has grown exponentially in recent years: information overload \u2014 with artificial intelligence as its secret weapon.<\/p>\n Maven fuses satellite imagery, drone feeds, radar data and signals intelligence into a single interface, then classifies targets, recommends weapons systems and generates strike packages in near real time, compressing kill-chain reasoning and decision making into the fastest timelines ever seen on the battlefield.<\/p>\n And it uses\u00a0Anthropic\u2019s Claude AI model<\/a>, embedded in its system, to semi-autonomously rank targets by strategic importance, drafting automated legal justifications for each strike along the way.<\/p>\n The software generated hundreds of strike coordinates in the first 24 hours of the Iran campaign, enabling the U.S. to hit more than 1,000 targets in the first 24 hours of the war, according to\u00a0The Washington Post<\/a>.<\/p>\n After sources briefed on preliminary findings told\u00a0CNN<\/a>\u00a0that U.S. Central Command had created targeting coordinates using outdated intelligence provided by the Defense Intelligence Agency that had not been updated to reflect the school\u2019s presence, one question became central to the inquiries: \u201cIf so, did a human verify the accuracy of this target?\u201d they asked.<\/p>\n They are still waiting for an official explanation.<\/p>\n Ukrainian drone operators who build and deploy semi-autonomous targeting systems on the front line told Military Times they recognized the likely culprit immediately.<\/p>\n Ihor Matviyuk, the director of\u00a0Aero Center<\/a>, a Ukrainian drone company that builds and deploys semi-autonomous drones on the front lines of the war with Russia, said he can imagine exactly how the failure happened.<\/p>\n Although he has no inside knowledge of the Minab strike specifically, earlier this month he said that it bears the hallmarks of a targeting failure \u2014 not an AI malfunction.<\/p>\n \u201cIt was almost definitely a strike on the [given] coordinates,\u201d Matviyuk told Military Times. \u201cThe main problem was not the AI \u2014 it was how close the military object was to the school.\u201d<\/p>\n Last week, former military officials speaking to\u00a0Semafor<\/a>\u00a0confirmed Matviyuk\u2019s early assessment: \u201cHumans \u2014 not AI \u2014 are to blame” for the school strike, they said, pointing to stale human-curated data fed to the Pentagon\u2019s Maven targeting platform.<\/p>\n Matviyuk recognized the pattern because he\u2019s had to decide how much AI to use in his own semi-autonomous weapon systems again and again as drone warfare and software capabilities have rapidly evolved on Ukraine\u2019s battlefield.<\/p>\n \u201cAutomatic targeting allows us to capture less than half of the targets, not more,\u201d Matviyuk said. \u201cBecause they are all still camouflaged.\u201d<\/p>\n The Defense Department\u2019s own data bears that out. Maven can correctly identify objects at roughly 60% accuracy overall \u2014 compared with 84% for human analysts.<\/p>\n But that rate drops below 30% in adverse conditions, such as bad weather or poor visibility, according to Pentagon data published in a 2024\u00a0Bloomberg<\/a>\u00a0report.<\/p>\n The risk of \u201ccollateral damage,\u201d as the strike on the Minab school might be categorized in military terminology, is too high \u2014 that is why Aero Center and every other Ukrainian drone company that spoke with Military Times says they always leave the final strike decision to a human operator.<\/p>\n \u201cThe direct impact is always carried out by the operator\u2019s command,\u201d Matviyuk said, \u201cto prevent civilians from getting under the blow.\u201d<\/p>\n In 2021, an experimental U.S. Air Force targeting AI scored roughly 25% accuracy in real conditions, despite rating its own confidence at 90%, then-Maj. Gen. Daniel Simpson, the Air Force\u2019s assistant deputy chief of staff for intelligence, surveillance, and reconnaissance, told\u00a0Defense One<\/a>.<\/p>\n \u201cIt was confidently wrong,\u201d Simpson said, summing up the program\u2019s problems. \u201cAnd that\u2019s not the algorithm\u2019s fault. It\u2019s because we fed it the wrong training data.\u201d<\/p>\n The situation is not expected to improve. Last month, Hegseth slashed the Civilian Protection Center of Excellence workforce by approximately 90% and cut CENTCOM\u2019s civilian casualty assessment team from 10 to one,\u00a0Politico<\/a>\u00a0reported.<\/p>\n Then, after leaving a skeleton staff to oversee the guardrails of the biggest expansion of AI in the military, Deputy Secretary of Defense Steve Feinberg signed a memo earlier this month formalizing AI\u2019s role in military decision making \u2014 designating Maven an official program of record and pushing adoption across all U.S. military branches by September,\u00a0Reuters<\/a>\u00a0reported on Friday.<\/p>\n Ukrainian weapon makers like Matviyuk are not shying away from giving AI more autonomy, but they\u2019re using it strategically.<\/p>\n Autonomous targeting is effective for \u201cmassive offensive operations, where targets are not camouflaged,\u201d he said, a description that may fit Iran\u2019s fixed military installations, which are far less concealed than most positions on the Ukrainian front.<\/p>\n \u201cWe support the idea of using the human element less and less in the drone operator job,\u201d Matviyuk said. \u201cAutonomy, autonomous elements of drones \u2014 that\u2019s the stuff we are working on.\u201d<\/p>\n The problem, in his view, was not that the Pentagon used AI. It was that the data behind the target had not been updated since a girls\u2019 school replaced a military headquarters on the same coordinates \u2014 and the people whose job it was to verify that data had already been cut from the chain.<\/p>\n AI systems are only as reliable as the people who build, feed and oversee them, Matviyuk emphasized.<\/p>\n When the human link fails, whether through bad data, gutted oversight or compressed timelines \u2014 the machine will continue to execute the error with precision.<\/p>\n Former CENTCOM director of intelligence,\u00a0Lt. Gen. Karen Gibson<\/a>, was unequivocal about where accountability for lethal strikes lies, regardless of weapon autonomy, at a\u00a0Center for Strategic and International Studies panel<\/a>\u00a0last week.<\/p>\n \u201cI will always come back to the fundamental principle of human responsibility and accountability,\u201d she said. \u201cA commander somewhere will ultimately be held responsible \u2014 not a machine or a software engineer.\u201d<\/p>\n![\"\"](\"https:\/\/www.fie.undef.edu.ar\/ceptm\/wp-content\/uploads\/2026\/03\/TTVPOBVBIBGH7KUE7DJUJPCJ6M.jpg\")