"Three clicks convert a data point on the map into a formal detection and move it into a targeting pipeline. These targets then move through columns representing different decision-making processes and rules of engagement. The system recommends how to strike each target – which aircraft, drone or missile to use, which weapon to pair with it – what the military calls a “course of action”. The officer selects from the ranked options, and the system, depending on who is using it, either sends the target package to an officer for approval or moves it to execution."<p>----------------<p>Maven is a tool for use in the middle of a war. When both sides are firing, minutes saved can mean lives saved for your side. Those lives, at least partly, balance the risks of hitting a bad target.<p>This was not a strike made in the middle of a war. If Maven was used in the strike that took out a school, it was being used as part of a sneak attack. Nobody was shooting back while this was being planned. Minutes saved were not lives saved. There should have been a priority placed on getting the targets <i>right</i>. Humans should have been double and triple checking every target by other means. This clearly didn't happen. The school was obviously a school that even had its own website. Humans would have spotted this if they had done more than make their three clicks and move on to the next target.<p>Whoever made the choice to use Maven to plan a sneak attack <i>without</i> careful checking made an unforced error when they had all the time in the world to prevent it. Whether it was overconfidence in their tools or a complete disregard for the lives of civilians that caused this lapse, they are <i>directly responsible</i> for the deaths of those little girls. I sincerely hope there are (although I doubt there will be) consequences for this person beyond taking that guilt to their grave.
I recommend looking closely at the New York Times analysis. There were factors that might have mitigated this as a strike target, but it also really did look like a part of the compound (and it originally was!). Yes, with hindsight, we can definitively know, and with sufficient time each target could probably have been positively ID'd, but there was precisely one mis-strike in 1000s of sorties, so this already is a low error rate. TFA discusses 50 specific strikes all of which missed via automated analysis. That doesn't seem the same.<p>I don't disagree there. But this is not a case of hallucination, and an existing website is a signal, not a determinant, of the real situation on the ground.
However, you have made a very, very strong assumption that these targets were not carefully evaluated. One that does not seem to be present in TFA or any analysis that I've read. In fact, the article itself quotes those in the know who <i>believe this should have been eliminated as a target</i>.
So I read the entire TFA, where do you see “quotes [from] those in the know who believe this should have been eliminated as a target”? I saw no such quotes about the school in TFA. Maybe I missed it.<p>> there was precisely one mis-strike in 1000s of sorties<p>How did you verify this? Because I’ll remind you, the U.S. administration denied responsibility for some time before owning up to this due to public pressure. Absent public pressure, I guess we would’ve had zero mis-strikes.<p>> so this already is a low error rate<p>As a father of similarly aged daughters, I can’t express enough how grotesque and disturbing the term “error rate” is here.<p>We targeted and killed young children. Plain and simple.<p>> However, you have made a very, very strong assumption that these targets were not carefully evaluated.<p>Let’s take the opposing assumption that this target was carefully evaluated then. Please reason through the implications now?
How many American schoolchildren have Iran killed in the last 25 years? How many Iranian schoolchildren have America killed?<p>Where's your moral justification for this war of choice if "oops, 137 dead kids is a normal expected outcome"?
Iran's regime is directly responsible for untold war crimes and suffering through their proxies. You can say the same about the US, which is why I dont think it would be immoral for Iran to launch a bunch of rockets at the US. 1/1000 being mistargeted isnt unexpected. Accidentally killing a bunch of kids would likely be worth it, morally speaking, if it led to the destruction of the Iranian regime.
If we had all military bases next to elementary schools, things might be different.
There are plenty of military bases next to elementary schools in the US.<p>Where do you think the kids of soldiers go to school?
The us has over 150 elementary schools on military bases. If you use a more colloquial definition of military base, many many national guard armories are on the same block as elementary schools or even right next to them.<p>Can you cite anything that says all iranian military bases are next to elementary schools? If they are on ALL bases, that makes hitting an elementary school on base less forgivable, not more, because if its a fact of every iranian military base, it's a lot harder to claim good intelligence and also that they didn't check that the part of base being bombed was the school.<p>Also, how is that relevant?
We do. Grocery stores (commissaries) and residential units as well.
The New York Times are the same people who spread the lie about Iraq having WMDs, they are not credible, and in fact have been proven to be incredibly biased when it comes to wars in the Middle East.<p>Israel and the US targeted many schools in Gaza. They killed tens of thousands of children. This strike was clearly intentional and very much in line with all other Zionist actions.
For someone that interested in precision of supporting claims with evidence, you make pretty ridiculous and completely unsupported claims yourself, like "there was precisely one mis-strike in 1000s of sorties".
>>I recommend looking closely at the New York Times analysis. There were factors that might have mitigated this as a strike target, but it also really did look like a part of the compound (and it originally was!).<p>What a ridiculous take. What does "originally was" mean? Maybe you wanna say "previously was"? That building was converted to a school 10 years ago! The intelligence they relied on is 10 years old!!!!! It's recklessness and stupidity dressed as bravery and courage.
What are you doing?
> Yes, with hindsight, we can definitively know, and with sufficient time each target could probably have been positively ID'd, but there was precisely one mis-strike in 1000s of sorties, so this already is a low error rate.<p>This is giving them too much credit.<p>Hegseth has already shown himself to entirely disregard the notion of War Crime, <i>even by the US military's own already controversial standards</i>. The double strike on the boats in the caribbean are literally the textbook example in US military textbooks of what not to do, and that it is a warcrime.<p>This was no mistake. It was the obvious outcome of a pattern of reckless action.
Wouldn't have been looking for targets if senile old fucks looking to deflect from their personal liabilities hadn't started shooting.<p>AI didn't do shit here. Stupid people built the AI and the weapons and applied them. Any other argument is intentional obfuscation.<p>You all are falling for propaganda.
I couldn't find a web site for the school when I searched for one and I also noticed that while schools are generally marked on Google Maps in Iran this school was not. Both are IMO not really relevant or reliable sources of targeting data anyways. I found very little evidence searching online for the school but I did find something that looked like a blog about a school trip. Again though the Internet is not a reliable source of data for targeting - should be obvious.<p>The main way targets should/would be selected is by direct intelligence. E.g. the targets should be identified through satellite or other observations. It's hard to imagine that a building that has operated for some length of time as a school would not have patterns that are visible from satellite vs. military facilities with different patterns. You also don't just randomly attack structures in this sort of surprise attack, you're presumably aiming for some specific people or equipment with some priority/military goal in mind, so you really want to have observed the targets and patterns and have up to date information on their usage.<p>I think what likely happened here is that the entire base was the "unit" of targeting and the mistake was in identifying which buildings were part of the base. In the satellite view the military buildings and the school look very similar (since the building as I understand it used to be part of the base but was repurposed as a school).<p>It's not true that whoever made the error had all the time in the world. Presumably once the order was given there was time pressure given that the strike was to be timed with the other intelligence.<p>In theory the US military should/is supposed to have good processes around this stuff. So we are told. Obviously failed in this case. It is a tragedy.
I agree with your overall sentiment, but how realistic is it? Israel/US says they've been hitting thousands of targets (so reality might mean ~hundreds, still a lot), how are they supposed to verify this at all?<p>> Humans should have been double and triple checking every target by other means.<p>How practically would this happen? The US/Israel don't want people on the ground, and people on the ground is exactly the only way you can actually verify stuff like this, not every place in the world is on Google Maps or have a web presence at all, so the only realistic way to verify this would be to visually inspect it in person, something neither parties who started this war want to do.<p>Even better, don't make attacks against other soverign nations that don't pose an immediately and critical threat to you, and this whole conflict could have been avoided in the first place.<p>But no, the president has to be involved in some sort of child-trafficking scheme, so pulling the country into a war seemed preferable to being held responsible, and now we're here, arguing about fucking details that don't matter.
The school literally had its own website. If the AI involved was as smart as the media hype machine makes them out to be, it would have found the website and marked it as a non-target. It never even would have made it to human review.
In this case, they would have discovered it was a school with a Google search, basically. There’s no excuse.
I'm pretty sure this is the school that was on the corner of a military base, and the school building hit was previously part of the military base.
Or the vast satellite network we run. Pretty easy to see it's school children going in and out of the area.
I agree with everything you said - but it's also the case that a set of parameters were created that, instead of requiring multi-person validation of target validity and provenance, prioritized speed to provide decision makers with options.<p>This certainly doesn't absolve the person implementing those parameters, but it is equally the responsibility of the very top of the decision-making structure.
Was the Iran War Caused by AI Psychosis?<p><a href="https://houseofsaud.com/iran-war-ai-psychosis-sycophancy-rlhf/" rel="nofollow">https://houseofsaud.com/iran-war-ai-psychosis-sycophancy-rlh...</a>
Its not a war crime if the AI does it?
"the question that organised the coverage was whether Claude, a chatbot made by Anthropic, had selected the school as a target."<p>This article is the first I have seen mention of Claude in relation to this specific incident. There's been plenty of talk about AI use in warfare in general but in the case of this school most of the coverage I have seen suggested outdated information and procedures not properly followed.
It's definitely been reported before that Claude was used for Iran attacks, at the beginning of March or earlier:<p><a href="https://www.theguardian.com/technology/2026/mar/01/claude-anthropic-iran-strikes-us-military" rel="nofollow">https://www.theguardian.com/technology/2026/mar/01/claude-an...</a><p>Edit: Also, <a href="https://www.washingtonpost.com/technology/2026/03/04/anthropic-ai-iran-campaign/" rel="nofollow">https://www.washingtonpost.com/technology/2026/03/04/anthrop...</a>
"The U.S. used Anthropic's Claude to support Operation Epic Fury against Iran yesterday, sources familiar with the Pentagon's operations tell Axios."<p>OK. The US probably also used telephones and Diet Coke.<p>Nothing cited said that Claude was selecting targets or informing target selection.
I have heard the claim everywhere.
there is a lot of confusion about all this stuff<p>you, today, can use Claude in Amazon Bedrock, and the way that works is, if you want it to be this way: the piece of code and model weights and whatever other artifacts are involved, they are run on Bedrock. Bedrock is not a facade against Claude's token-based-billing RESTful API, where Anthropic runs its own stuff. In the strictest sense, Bedrock can be used as a facade over lower level Amazon services that obey non-engineering, real world concerns like geographic boundaries / physical boundaries, like which physical data center hardware is connected by what where / jurisdictional boundaries, whatever. It's multi-tenancy in the sense that Amazon has multiple customers, but it's not multi-tenancy in the sense that, because you want to pay for these requirements, Amazon has sorted out how to run the Claude model weights, as though it were an open-weights model you downloaded off Hugging Face, without giving you the weights, but letting you satisfy all these other IP and jurisdictional and non-technical requirements that you are willing to pay for, in a way that Anthropic has also agreed.<p>This is what the dispute with the Pentagon is about, and what people mean when they say Claude is used in government (it is used in Elsa for the FDA for example too). Anthropic doesn't have telemetry, like the prompts, in this agreement, so they have the contract that says what you can and cannot use the model for, but they cannot prove how you use the model, which of course they can if you used their RESTful API service. They can't "just" paraphrase your user data and train on it, like they do on the RESTful API service. There are reasons people want this arrangement ($$$).<p>The vendor (Palantir) can use, whatever model it wants right? It chose Claude via "Bedrock." I don't know if they use Claude via Bedrock. Ask them. But that's what they are essentially saying, that's what this is about. Palantir could use Qwen3 and run it on datacenter hardware. Do you understand? It matters, but it also doesn't matter.<p>It's a bunch of red herrings in my opinion, and this sort of stuff being a red herring is what the article is mostly about.
Worth mentioning that the author wrote about this first on his substack: <a href="https://artificialbureaucracy.substack.com/p/kill-chain" rel="nofollow">https://artificialbureaucracy.substack.com/p/kill-chain</a>
Turning a military building into a girl's school, and then having this school right next to other military buildings - is this something that happens often? Or were there ulterior motives behind it?
Yes, the US has 160 schools on military bases:<p><a href="https://www.militaryonesource.mil/education-employment/for-children-youth/what-schools-are-available-to-children-on-military-installations/" rel="nofollow">https://www.militaryonesource.mil/education-employment/for-c...</a>
Really fascinating article. Bits of bias here and there, like "The US military has been trying to close the gap between seeing something and destroying it for as long as that gap has existed" -- you can respond to seeing and understanding something without destroying it -- but it underscores, to me at least, how much denser the "fog of war" has become. The fog of media reporting in general. Those first few paragraphs felt like a breath of fresh air.
Before it was the gods, then God, then Nature, and now AI. Human beings really have a fundamental issue with accepting responsibility for their actions.<p>From a certain angle, the entire industrial and computer age looks like a massive effort to remove all responsibility for our actions, permanently.
Interesting article. Seems like AI-washing isn't just for layoffs anymore.
When AI gets something wrong, it's the operator's fault, IMO.
You can't have a serious discussion of this bombing without addressing the information warfare component. To this day <i>we don't know what actually happened</i>. Between the general public and the facts, there are many middlemen, all with their own distorting factor: the IRGC; the US government; western press outlets such as the Guardian; and the people quoted by the press.<p>IRGC is making claims that no other party can verify first-hand. Everything from the number of explosions, the extent of the physical damage, the number of wounded and dead, the number of <i>civilians</i> wounded and dead - these are all unverified claims and should be treated as such. Not only is the IRGC obviously biased and incentivized to maximize media pressure on the US and Israel: they are known for information warfare of exactly this nature. To take their statements at face value, and present them as established facts in the opening paragraph, as this article does, is journalistic malpractice.<p>Again, <i>the basic facts on the ground are not known</i>, yes all parties are projecting narratives with a certainty that we should all be suspicious of.<p>Without this stable foundation of knowing what actually happened, and why, the very premise of this article collapses on itself.<p>EDIT: the flurry of responses to this post illustrate the problem. It's difficult to even have a respectful, fact-driven discussion on this topic, because everyone is tempted (and encouraged) to rush to their political battle stations. Nobody wants to discuss information warfare, because they're too busy engaging in it. I think that's worrying and problematic. No matter which "side" you're on, it should be possible to distinguish what is known and what is not; and implementing basic information hygiene. Or do you think you are uniquely immune to disinformation?
You're not wrong but what we can tell from open sources is:<p>- The building does seem to have actually been a school and "detached" from the rest of the military complex.<p>- The school the Iranians claim it was does seem to exist even if it's not 100% clear that's the identical location.<p>- At the time of the attack school would have been in session.<p>- The signature of the attack seems similar between all the buildings attacked and we have footage showing a Tomahawk hitting the area.<p>Another thing we can tell is that the US has to know the truth here and isn't coming out with an official statement.<p>And I'm saying this as someone who thinks the Iranian regime is evil, needs to be struck down, was trying to acquire nuclear weapons etc.<p>As to the numbers I agree they are to be treated with suspicion. The Iranians are obviously motivated to lie, inflate them, and treat all casualties as civilians. But we can still try and estimate given the size of the building what would be the number of students. We can also estimate the outcome of the missile hitting the building and correlate with the photos and satellite imagery, and until we have better data use those estimates.
I agree with all of that. My worry is that the Guardian article is not doing <i>any of it</i>, and in fact is damaging the framework for even having such a conversation.. Instead they are repeating IRGC statements without attribution, and establishing them as background truth in the first paragraph. Then building an entire article on that flawed premise. Essentially, their article exists in the narrative universe create by the IRGC. I find that incredibly worrying.
Everyone acknowledges that the US killed a whole bunch of kids, including the US
Holy gaslighting bootlicking
I think its fair to treat things that the Trump administration and the Iranian military agree on as facts. If they were distortions that favored one side, we would see pushback from the other. <i>Maybe</i> there are distortions that somehow benefit both of these parties, but it seems unlikely. At minimum, then, this was a school, the Americans bombed it, and children died as a result.
No. The only thing that the US government and IRGC agree on, at the moment, is that there was an explosion at the site of the school.<p>The US did NOT confirm that they are responsible for the bombing, or that children (or anyone) died as a result. This is a verifiable fact.<p>So, applying your own principle: the only thing you should treat as fact, is that there was an explosion at a school.
> To this day we don't know what actually happened.<p>I feel like we know enough already. A school was bombed, the ones who did it sucks big time and should be held responsible. Currently, the US and Israel is waging a war against Iran, and one of them dropped the bomb(s), unless suddenly Iran got their hands on American weapons, then that needs to be investigated too, because someone surely dropped the ball at that point.<p>The basics remain the same, investigations have to be launched to figure out where exactly in the chain of command, someone made a mistake, and then hold that person(s) responsible for their fuck up.<p>Have those investigations been launched?
I think it's likely that the explosion was caused by a US strike. But we don't actually know for sure that that's what happened - the US government has not confirmed it.<p>We also don't know anything about casualties - we only have the IRGC statements, and they are not reliable.<p>> <i>Have those investigations been launched?</i><p>Yes, according to the US government, an investigation is underway. But its starting point is determining <i>what caused the explosion</i>.
How long does it take to look at the coordinates programmed into the cruise missiles? Or to review existing satellite imagery for the location and other intelligence sources?<p>If this was a school (which seems likely at this point) and if this was a US TLAM that hit it (which also seems likely at this point) then we should expect a lot of casualties when it's hit during school time (which also seems likely). And yes, we shouldn't trust what the IRGC is saying.<p>I think I'm on your side but in this case the correct course of action for the US would have been to quickly own up to the mistake. There is really not a lot of ambiguity here. This doesn't seem to be a case like "shots were fired from the school window" or some sort of dual use with IRGC having offices in the school. If there was a reason for the targeting then presumably we'd have a statement about it already.<p>Mistakes can be made and are always made in war. Leaving this open like this is damaging to the war effort.
You are the one engaging in "information warfare", intentionally trying to spread doubt about an event that was confirmed by both Iran and US. What does it feel like to deny the murder of 150+ children out of nationalistic pride? Do you simply have no conscience? No sense of guilt, no concept of morality?
I feel like an intellectual god to have been gifted the brain power to recognize that 150 kids being killed is a awful tragedy, and that converting a building on a military base to a school is recklessly stupid and borderline purposely done as a trap. It's like letting your child play in the road at night, and then being upset when a drunk driver hits them.<p>Anyone can look at the satellite images from the bombing and see how ridiculous whatever Iran was doing was.[1]<p>[1]<a href="https://npr.brightspotcdn.com/dims3/default/strip/false/crop/1600x900+0+63/resize/1400/quality/85/format/jpeg/?url=http%3A%2F%2Fnpr-brightspot.s3.amazonaws.com%2Fc6%2F45%2F6674da344d7e85243a94d6630f4e%2Fwide-after.jpg" rel="nofollow">https://npr.brightspotcdn.com/dims3/default/strip/false/crop...</a>
"I understand that the officer killed your unarmed teen son. But you have to understand, in the dark, he appeared to be reaching for a weapon, and the officer feared for his life."<p>"It's a tragedy that she was raped. But you have to understand, the way she was dressed, she clearly wanted it, she was sending mixed signals, you see."<p>Anyway. Here's a preschool right next to a military base, it took me about 3 minutes of scrolling around on google maps to find this.<p><a href="https://maps.app.goo.gl/2TP32tYqRZxthSFF8" rel="nofollow">https://maps.app.goo.gl/2TP32tYqRZxthSFF8</a>
I feel very bad for the children and their relatives. What happened is really terrible.<p>I wish there was the same level of rigour and energy applied to investigating the 40,000 deaths in early January. There are countless videos online.<p>I simply don’t understand why 150 people receive so much attention while 40,000 don’t.<p>This saddens me because it feels like the focus is on who was responsible rather than who lost their lives.
[flagged]
Something that a lot of tech people, especially in Silicon Valley, seem to want to forget, is that at every level you still have people making decisions. AI is suggesting but someone, somewhere, still has to make the decision to act on that suggestion.<p>It's still people doing people things.
The immediate concern isn't really fully autonomous systems, it's that the nature and design of recommender/suggestion systems prompt humans to sleepwalk through their responsibilities.
<a href="https://en.wikipedia.org/wiki/Computer_says_no" rel="nofollow">https://en.wikipedia.org/wiki/Computer_says_no</a>
You know how that was done with a Tomahawk<p>They've now burnt though almost ONE THOUSAND of those<p>They cost $4 million each, so that's another $4 BILLION that has to be replaced too<p>Imagine several more months of that or even through 2029
The Royal United Services Institute (RUSI) has an updated tally on defensive and offensive munition expenditures. It's likely not 100% accurate due to the sensitive nature of those figures.<p>> 11,294 munitions in the first 16 days of the conflict, at a cost of approximately $26 billion.<p>Several detailed tables are in the link below.<p><a href="https://www.rusi.org/explore-our-research/publications/commentary/over-11000-munitions-16-days-iran-war-command-reload-governs-endurance" rel="nofollow">https://www.rusi.org/explore-our-research/publications/comme...</a>
We'll run out long before 2029. The 850 fired so far is about a quarter of the entire supply.<p><a href="https://www.reuters.com/business/aerospace-defense/us-uses-hundreds-tomahawk-missiles-iran-alarming-some-pentagon-wapo-reports-2026-03-27/" rel="nofollow">https://www.reuters.com/business/aerospace-defense/us-uses-h...</a>
So far, they are not funded to do this for that long. They have floated a $200B bill to congress, which made national news coverage. It would start a huge, prolonged fight over the war and actually force them to ask permission from congress to fight it (barring totally disregarding the constitution which is still a possibility).<p>Unfortunately I can very well imagine several more months and years of this. We are still fighting a forever war that started in 2001. This is all a generation of Americans will know, and that is sad.
It’s a tale as old as time: start a war to support the military industrial complex. Imagine a $4 billion investment into public transportation or parks. Every 10 years we can invest into a new city instead of bombing some kids overseas (whose siblings, fueled by hatred, then commit terror attacks on the west).
Don’t worry. The Saudis and UAE will happily pay all costs of the war.
[dead]
[flagged]
[flagged]
The WTC complex had defense department offices: <a href="https://www.govexec.com/federal-news/2001/09/federal-agencies-located-in-the-world-trade-center-complex/9931/" rel="nofollow">https://www.govexec.com/federal-news/2001/09/federal-agencie...</a><p>By your logic it's the federal government's fault those 3000 people died on 9/11, they were being used as human shields.
> This was a choice to use children as human shields<p>Perhaps we should have, you know, just not bombed that particular fucking site until the end of the fucking school day if it was such a vital target. God forbid we act like a vaguely intelligent country, instead of drunkly screaming "maximum lethality" at every conceivable opportunity.
They (the Americans) should have also marked the schools on said military maps of theirs, and hence they could have made a value judgment of "is it worth killing some IRGC men in the middle of nowhere vs. the international backslash of killing school-going children?". It looks like they most probably didn't do that, probably because their "advanced" AI systems didn't bother with marking schools on their military maps.
American bases in europe have schools on them.
Fair targets?
Israel and the US are bombing lots of schools and hospitals and civilian infrastructure, this is not the only case. This is intentional genocide, not a software/organizational/human error.
Check out this example: <a href="https://www.bbc.com/news/articles/cjw4egp7lwno" rel="nofollow">https://www.bbc.com/news/articles/cjw4egp7lwno</a><p>"As we pass through Khan Shaykhun, we come across a street painted in the colours of the Iranian flag. It leads to a school building that was being used as an Iranian headquarters." "On the wall at the entrance of the toilets, slogans read: "Down with Israel" and "Down with the USA".<p>It was evident that these headquarters were also evacuated at short notice. We found documents classified as "highly sensitive"."<p>This is a BBC reporter reporting from Syria after the fall of Assad.<p>It is strategy for the IRGC and Hamas to operate from civilian infrastructure like schools to gain immunity. That's what's "not a human error".
Sufficiently advanced negligence is indistinguishable from malice.<p>This is not to say that this administration is definitely <i>not</i> targeting civilians or infrastructure on purpose; just that the end result, and the moral culpability, are the same in either case.
>The targeting for Operation Epic Fury ran on a system called Maven. Nobody was arguing about Maven.<p>Would it be poor taste to make joke about gradle being superior here? The dad in me <i>really</i> wants to make that joke...
The Guardian carrying water for the AI industry. The distinction between Maven and Claude is futile. We get that Maven is Palantir, but it integrates Claude:<p><a href="https://www.reuters.com/technology/palantir-faces-challenge-remove-anthropic-pentagons-ai-software-2026-03-04/" rel="nofollow">https://www.reuters.com/technology/palantir-faces-challenge-...</a><p>Going into a generic rant about anti-AI people after missing sources and believing the Department of War is just extremely poor journalism from the newspaper that destroyed evidence after a command from GCHQ.<p>I hope this is a single "journalist" and that the Guardian has not been bought.
I assume you actually read the article and didn't just post this after a quick skim, yes? Because saying this:<p>> The distinction between Maven and Claude is futile<p>Doesn't make any sense at all when you read the article and understand what Claude actually does in this equation. From the article:<p>> Neither Claude nor any other LLMs detects targets, processes radar, fuses sensor data or pairs weapons to targets. LLMs are late additions to Palantir’s ecosystem. In late 2024, years after the core system was operational, Palantir added an LLM layer – this is where Claude sits – that lets analysts search and summarise intelligence reports in plain English. But the language model was never what mattered about this system.<p>The <i>whole point</i> here is that whether an LLM is involved or not is immaterial to the system as a whole, and it's a <i>disservice</i> to the public to focus on LLMs here.
The article you're responding to is making specific operational claims about Claude's (basically non-) relevance. I'd be interested to hear if you're directionally correct, but forgive me if I need more details from your counterargument than "but it integrates Claude".
This is not a correct take at all given the contents of the article.
Better than carrying water for people who blame inanimate tools for their own personal and professional failures.
WaPO writes that Claude selected targets:<p><a href="https://www.washingtonpost.com/technology/2026/03/04/anthropic-ai-iran-campaign/" rel="nofollow">https://www.washingtonpost.com/technology/2026/03/04/anthrop...</a><p>This unknown Guardian contributor writes a missive against "Luddites" while using the typical AI booster arguments that always turn around anti AI arguments.<p>Just like two five year olds: "You have a big nose." "No, <i>you</i> have a big nose."<p>We learn from this clown that anti AI people suffer from AI psychosis because they are reading WaPo and Reuters.
Both the Washington Post and the Guardian articles agree that the system used here was Maven.<p>The key sentence in that Washington Post article appears to be:<p>> The Pentagon began to integrate Anthropic’s Claude chatbot into Maven in late 2024, according to public announcements.<p>As far as I can tell this is the public announcement - a press release from November 2024: <a href="https://www.businesswire.com/news/home/20241107699415/en/Anthropic-and-Palantir-Partner-to-Bring-Claude-AI-Models-to-AWS-for-U.S.-Government-Intelligence-and-Defense-Operations" rel="nofollow">https://www.businesswire.com/news/home/20241107699415/en/Ant...</a><p>> Anthropic and Palantir Technologies Inc. (NYSE: PLTR) today announced a partnership with Amazon Web Services (AWS) to provide U.S. intelligence and defense agencies access to the Claude 3 and 3.5 family of models on AWS. This partnership allows for an integrated suite of technology to operationalize the use of Claude within Palantir’s AI Platform (AIP) while leveraging the security, agility, flexibility, and sustainability benefits provided by AWS.