# Inside the US ‘Kill Chain’ That Destroyed an Iranian School | Patrick Galey **By:** Patrick Galey **Published:** 2026-09-19T12:12:55.218Z **Source:** [LinkedIn](https://www.linkedin.com/posts/patrick-galey-8158b01b_amazing-work-from-bloomberg-news-investigating-share-7507049080536104961-ydi4) --- # Patrick Galey’s Post [Patrick Galey](https://uk.linkedin.com/in/patrick-galey-8158b01b?trk=public_post_feed-actor-name) Amazing work from [Bloomberg News](https://www.linkedin.com/company/bloomberg-news?trk=public_post-text) investigating how the US military used artificial intelligence developed by Palantir and using Anthropic’s Claude to murder 150 schoolgirls and teachers at a school in southern Iran in February. “Inside Centcom, which conducted the US attack, some personnel relied too much on the artificial intelligence embedded in Maven Smart System, the officials said. Maven, which is made by Palantir Technologies Inc., allows users to view and coordinate complex military operations — from targeting to command and control — using AI-enabled software that fuses more than 150 different data inputs into a coherent set of information to inform commanders’ decisions. “Some Centcom personnel expected Maven to flag stale information or inconsistencies in the underlying intelligence assembled for potential targets, the officials said. It’s not clear why they had such expectations.” I hear a lot about the potential use cases of hyperscaled AI products. Well, how about the actual, real-world, harrowing and hard-lived and died use cases of them: accelerate the mechanised slaughter of foreign civilians. [https://lnkd.in/eA\_SUkPd](https://www.linkedin.com/redir/redirect?url=https%3A%2F%2Flnkd%2Ein%2FeA_SUkPd&urlhash=PyGv&trk=public_post-text) [ Inside the US ‘Kill Chain’ That Destroyed an Iranian School bloomberg.com ](https://www.linkedin.com/redir/redirect?url=https%3A%2F%2Fwww%2Ebloomberg%2Ecom%2Fgraphics%2F2026-iran-school-attack%2F&urlhash=WAcH&trk=public_post_feed-article-content) Even better than that, it allows us to ignore the names of \*people\* who committed this war crime, and instead talk about the AI like it's some sort of technical issue to be fixed. I'm just waiting for "I didn't launch the nuke, the AI did it". This is the real danger of AI. Not some "Terminator" tech bro wet dream, not even "WOPR", but the removal of accountability from sociopathic leaders, making ever-increasing criminal acts feasible without censure. No doubt the actual versions of software used in the attack no longer exist, rendering the defence unfalsifiable. It's also the message of these "rogue AI broke into xxxx" claims. We are being conditioned to believe the myth of rogue AI. I have analyzed several of the legal issues raised in Minab and Lamerd around feasibility of precautions, AI-enabled warfare and civilian harm in a piece I wrote for [Opinio Juris](https://www.linkedin.com/company/opinio-juris-blog?trk=public_post_comment-text) here in light of the UN’s fact-finding mission report and related to this Bloomberg piece, in case of interest: [http://opiniojuris.org/2026/09/18/un-fact-finding-mission-on-iran-u-s-war-crimes-and-precautions-in-ai-enabled-warfare](http://opiniojuris.org/2026/09/18/un-fact-finding-mission-on-iran-u-s-war-crimes-and-precautions-in-ai-enabled-warfare?trk=public_post_comment-text)/ [Patrick Galey](https://uk.linkedin.com/in/patrick-galey-8158b01b?trk=public_post_comment-text) The most terrifying possibility is not that AI will make war more efficient-it is that it may make the killing of human beings look like a data problem. When an algorithm converts uncertainty into a targeting recommendation, a human life can be reduced to a probability, a confidence score, or a data point. And once the machine is treated as an authority, who answers when the algorithm is wrong? The machine cannot stand before the dead. This is the ultimate ethical red line: AI must never become the mechanism through which humans outsource not only decisions, but responsibility for killing. Fully automated "Lucifer Effect" collective psychosis... Seven social processes that grease the slippery slope of evil: • Carelessness • Dehumanisation • De-individuation • Lack of personal responsibility • Blind obedience to authority • Uncritical groupthink • Indifference and apathy The IRGC is now thought to have slaughtered hundreds of thousands of it's own citizens. If true, this would be the largest massacre in such a short period of time in history. Fascinating timing given calls for regulation over the past week+ from some of these same companies (e.g., possible preemptive PR campaign in advance of this news drop) – hope this latest info helps inform the UN policy decision on weaponized AI in Nov. It’s a pity though that the fact-finding mission report does the subject such a disservice. And who is accountable for these deaths? No one. My Lai -- gaming mangles the mind.  Cfbr