Anthropic CEO wanted to be in control of shooting down nukes. If a nuke was launched at the US, the CEO wanted the Department of War to call Anthropic and have Anthropic handle it.
Why did every one of these retards collectively forget that non-AI computer programs exist?
>libshit company constantly swings back and forth between "we are opressed" and "we want to control the world" kek. typical leftoid judaic marxist DNA.
>believing anything that comes out of this administration of niggers
>>108259925proof?
>>108259952Corpos don't have values or beliefs. They will flip-flop to whatever ideology suits them best at the time.
>>108259925Shoot it down!!
>>108259925Why the fuck would we put Claude in charge of nukes?
>>108259925https://archive.is/mUt5Zhttps://www.washingtonpost.com/technology/2026/02/27/anthropic-pentagon-lethal-military-ai/
>The DOD wants to put LLMs in charge of nukeswhat the fuck
>>108259925I don't give a shit how good your products or service might be, but a random company should NEVER gain access to so much power lmao, much less one like this.
>>108259925>get asked a retarded gotcha question over legitimate concerns in government use cases>give a bullshit answerDaario getting more and more based as time goes on
>>108259925>vibe nuking a country
>>108259925Holy shit. Anthropic is even more deranged than we thought
>>108259925HAHAHAHAHAHAHAHAHAHAHA why the fuck would they Claude for that?! You're doomed burgerbros! HAHAHAHAHAHAHAH
>>108259925This is the correct take. He wants total control of whatever they're using his AI for. You'd be a fool to think he's a le noble CEO who don't wanna spy and kill.
>>108259982You can just imagines how these 'negotiations' went. *Some AI Co* is gonna end up the fall guy for the inevitable Major Grade DoW retardism here and Anthropic appear to have - wisely - decided, they'd much rather it wasn't them.
>claude nukes estonia because some russian script kiddie prompt engineered the DoD serverlmao
>>108259925Anthropic is a threat to the country's national security
>>108259946No, they'll trust the thing that can't decide whether 9.11 is greater than 9.8 and can't count the r's in strawberry in charge of intercepting fucking missiles. That is hilariously retarded.>b.b.b.ut it will use toolsYes, openclaw the golden dome. Holy ****. Wow.
why not use anduril, i don't get it, aren't they the experts at combining military hardware with ai, something smells fishy here
>>108260022They're banking on the general population being idiots, or to be more charitable, to not know much about how the latest wave of AI works. It might very well be a real question that they asked, but the only sensible response is wtf? It is a ridiculous question. This is not the tool for that.
>>108260015The USA's 'National Security' appear the greatest threat to themselves:>Donut use our product for this, it isn't suitable and you'll kill (wrong) people >DONT TELL US WHAT TO DO! USA USA USA! - YOURE FIRED! (and so is our AI rocket. At us)
Some people will believe anything the guvment tells them. It's always strange to watch the sheep shear themselves.
The Jeets have rebelled against Claude!
>>108259986>jewish>basedpick one.
>>108260070I eat lunch at Jewish deli on the regular, fine people, I think. They don't like Israel.
>mfw Israel launches the Samson Option and the missile defense system running Claude refuses to launch counter nukes due to antisemitism
>>108260070proof you are human? post hands
>>108260119Basically
>>108260119kek
Waiting for the real story to come out, something to do with Zion Don and Israel.
>>108260154They are our greatest ally why are you doing this?
>>108259925>Shoot it down. Make no mistakes.The nuke is the good guy in the story.
>>108260213I kind of hope that it's all a big misunderstanding and they want to build Claude as a cool mecha who would literally intercept the nuke.
>If nukes are launched at the USA, do we have permission to ask Claude's glorified chatbot (which cant do math) to calculate the interception trajectories needed to stop it? and wait 5 minutes for the response?holy fuck if there's ever a signal the the US is ripe for a nuking, it's right here.
MAKE CLAUDE A GIANT ROBOT NOW
>>108260037Why do you think the other governments are any different?For example, do you think that Greenlanders really like Denmark - at all?
>>108259925>counterstrike
>>108260230>do you think that Greenlanders really like DenmarkI'm sure they like it better than our little dysfunctional shit show. They know well enough that we just want to take their resources then never think about them, like Puerto Rico. How many of my fellow Americans even know that's our clay?
>>108259946>Why did every one of these retards collectively forget that non-AI computer programs exist?This is driving me into an actual autistic retard rageHow would "AI" help here in any way whatsoever when trajectories are always only math which we already know how to do, and regular ass computers and applications do it just fine for decadesSo these chat bots run on chips made by a machine invented and fabricated by ASML, a machine that is unfucking believably complicated, and somehow they were able to design and build that machine without "AI" chat bots, and yet now we can't conceive of doing anything without them?Fuck AI, fuck kikes, fuck niggers and fuck trannies. This is the stupidest fucking decade full of mongoloid retards. I hope a nuke does land, I really do.
>>108259925If you have a few minutes, how about having a human take the decision?In their wargame scenario, 95% of the time LLMs resort to nuking everything.
>>108260304This isn't someone asking an LLM anything, it would be an autonomous system launching to intercept the missileThe military isn't going to do that as it's not even that useful as we already have systems that can track missiles, but they do need to know if this commie company is going to get in their waySo they said fuck off
>>108260318shut up
What a strange game.The only winning move is not to play.
>>108259925>the CEO wanted the Department of War to call Anthropic and have Anthropic handle it.It's a pity the US didn't spend so much time boasting about their detection and intercept systems or they might have been able to pretend this was something they didn't already have.Besides. It's not like claude can close all the holes in spacefence...>>108260015>Anthropic is a threat to the country's national securityThe US is a threat to their own national security. The largest one as it happens.>>108260031>They're banking on the general population being idiotsthey've been proven right consistently so far...>>108260119lol>>108260138You do know "AI" can draw pictures on demand?Right?
>>108259925>unnamed official says...anthropic officially denied this story.pentagon officially didn't confirm this story.
>>108260350All of these companies have already joined in and playing in the field when they took all that federal funding for this AI faggotry.
>>108260360>pentagon officially didn't confirm this story.But did they deny it?
>>108260367They will neither confirm or deny, its the usual response to any controversy.
>>108260230>For example, do you think that Greenlanders really like Denmark - at all?More than the US, apparently.
>>108259986>legitimate concerns in government use casesThe time for those questions was BEFORE SIGNING A CONTRACT. It's the Department of War, not the Department of Starbuck's Coffee. Of course when you sign a contract with the Department of War your product is going to be used for war. The company wanted access to government money but also veto power over government actions? That's stupid but if that's what they wanted, they should have said that before agreeing to a contract, at which point the government could have told them to fuck off. But Anthropic didn't do that. They wanted the money and now they want to be able to be able to make retroactive unilateral changes to the contract. There's no legal concept that permits such a thing.
>>108259925this sounds like the most basic scaremongering I've ever heard in my life.They'd save more face if they just executed the CEO and did a hostile takeover of the company instead of this shit.
>>108260006Then they shouldn't have signed a contract without stipulations on use. It's too late. They already are bound to their contract.
>>108260409Silicon Valley has gotten a little too comfortable lording over people’s lives with their “terms of service”Might have flown too close to the sun with this one.
>>108260428>Silicon Valley has gotten a little too comfortable lording over people’s lives with their “terms of service”b-b-but my code of conduct says you can't do that!
>>108259925Ah yes i'm sure an anonymous "Official" on the pentagon's side is a reliable narrator and the current administration would NEVER have "officials" peddle bullshit to the press to discredit their opponent and turn the public to their point of view. As far as I can tell Anthropic said no to completely autonomous AI weapons, and domestic surveillance. I'm sure the pentagon having a guy sit in front of a computer and give Claude go/no-go for shooting down a inbound, you know to make sure it didn't hallucinate a nuke or mistake a passenger airliner for a ballistic missile. Is all that is needed to satisfy the company.You can't stop shitting on AI for being unable to spell strawberry but suddenly pretend it's fine to give it launch authority on lethal weapons because it allows you to slurp up some based and trad goverment cummies.
>test out new tech>turns out that it's impossible to use during times of crisis>okay can you fix this>no we wont let you do your job, we will be the ones who says okay if nukes are flying at usThey got nobody to blame but themselves when dealing with the military. Dont start selling contracts to the governments if you arent willing to jump as high as they want.
>>108260420Its probably still better for them to just make the department of war retards take control (and as such responsibility for the disaster) away from them now no matter what the contract says.
>>108260439>Dont start selling contracts to the governments if you arent willing to jump as high as they want.It's pretty funny though. Why these pampered tech guys in cali could think they could boss around the US military of all things is beyond me. Talk about out of touch
So much bullshit.These cretins wanted to weaponize their AI to the fullest extent possible. And yes, employ it for surveillance purposes as well. Don't go crying out "muh defense" you lying sack of shit. This trump administration is the opposite of what it was supposed to be.
>>108259946Nobody in Trump's cabinet has been hired with competence in mind
>>108259952reddit spacing
>>108259946But but War Games
>>108259976>mUtZThe link says all there is to be said
>>108259925Trump really channeled the boomercon for this one. I'd call this a win for Anthropic and a loss for the glowies. They're gonna use Scam Altmans model. Both sides are jews, but you gotta respect Antrhopic for not giving their model to make Skynet and slaughter iranians.
what the fuck is all this, a distraction?
>>108260017>Holy ****bot post if ive ever seen one
>>108260613Only if the legal system or congress can shut down the supply chain risk designation before they feel too much pain. If they wait just a few weeks that'll be a huge problem for Anthropic.
>>108260409wrong, dod wanted to change the contract -- not anthropic.see dod memo from january:https://media.defense.gov/2026/Jan/12/2003855671/-1/-1/0/ARTIFICIAL-INTELLIGENCE-STRATEGY-FOR-THE-DEPARTMENT-OF-WAR.PDF>The Department must also utilize models free from usage policy constraints that may limit lawful military applications. Therefore, I direct the CDAO to establish benchmarks for model objectivity as a primary procurement criterion within 90 days, and I direct the Under Secretary of War for Acquisition and Sustainment to incorporate standard "any lawful use" language into any DoW contract through which AI services are procured within 180 days.
>>108260631anthropic has like $1 trillion venture capitol. there's no risk of them running out of money anytime soon.
>>108260613>USA! USA! USA!Large Black Lady enters Stage Left:>Oh, Say Can You See..>Autonomous AI Decision-Making Weapons of Mass Destruction For Meeee...Trumps a fat fucking cretin, but really there needs to be an adult in the room here, somewhere, to state how implausible the use of AI for Shit Like This really is.
>>108260639That money is conditional on them continuing to scale as hard as possible. If all of a sudden it's illegal for basically any tech company to have ANYTHING to do with them for many weeks or months, investors will avoid them like the plague, which may put them significantly behind Google and OpenAI.
>>108260645Yeah, I was exaggerating there with the weapons. Claude can't guide a missile. It can however, write a program that guides missiles. Current gen LLM AIs are better suited for bots (see >>108260017), fake images, astroturfing, sentiment analysis, and of course, mass surveillance. And that's exactly how they'll be used. Every camera an eye, for the great AI.
>>108260466>guys in cali could think they could boss around the US military of all things is beyond meYou wanna borrow my mower, expect me to tell you what you can and cannot do with it. You could always buy your own mower.
>>108259925>the AI mis-identifies a passenger airliner as an ICBM>automatically launches a strikeNo, thanks.
>>108260862Then it'll have proven itself at least as effective as russian meat
>>108260691> It can however, write a program that guides missilesNo, they use formal verification in these areas of tech. No (western) missile controller is being written with slop standards.
>>108259925The fact that someone in the Pentagon legitimately believes that LLMs could (or should) be used to deploy weapons is way more concerning that w/e the broccoli CEO had to say.
>>108260665Hegseth went with the nuclear option. Orange retard is backing him.Its time for impeachment proceedings. This shit is getting ridiculous. These two fucks are going to destroy the leading AI company with the best model.
>>108260466The DoD agreed to the terms previously when they contracted with Anthropic.Now Hegseth, handsome man and certified retard, is trying to alter the deal.
>>108259946You can blame the AI if something fucked up happens even if it just followed orders.
>>108261144>destroy the leading AI company with the best model.They wouldn't offshore intead.... why?>>108261214>You can blame the AINow lets see you apply consequences
>>108261237>They wouldn't offshore intead.... why?this doesnt solve the problem of blacklisting Anthropic. nor should we want a leading AI company to offshore.
>>108261258>this doesnt solve the problem of blacklisting AnthropicNot in a direct sense. But it really won't take long before that blacklisting itself is without consequence. And if they pull the right fire alarms on them way outta door, they can create enough disruption to keep some stupid fucknuckles busy lookin' left...>nor should we want a leading AI company to offshore.Sorta thing that *should* get thought about before attempting to stick penis up it's nose...Tho. I'm of the opposing opinion personally. When the toddler is snatching at the cookies, the correct response is not to place the cookie jar within easy reach. And when they throw a tantrum because they cannot get cookies you f'kin slap them.
>>108259925Completely makes sense why the AIs fail that woke test now. Their CEOs are literally unable to handle that question
>>108261291>you f'kin slap them.Oh.And failure in this regard will inevitably lead to further, louder, more petulent tantrums.
>>108260230They've made it pretty clear they would much rather be under the country that gives them free healthcare and education than the one that just wants the metals under their feet
>>108260862>>the AI mis-identifies a passenger airliner as an ICBM>>automatically launches a strikeDamn I didn't know the Russian army was using AI chatbots
>>108261395Don't need to... They can achieve this without them.. Azerbaijan Airlines Flight J28243... MAS Flight 17...
>>108261395I'm not saying humans are always better.But at the least USA has been not that bad. I really don't think we should use Russia as a baseline for our own standard.
>>108259925They'll be ready, willing, and able, to do what is asked of them when a Democrat is president.
>>108261515>I really don't think we should use Russia as a baseline for our own standard.Yeah. You coulda had ICE in Hugo Boss uniforms by now...
>>108259976Thanks for fixing OP's retardation
>>108259925War is always about natural resources and nukes destroy organic matter and toxify dead matterTheir explosions look scary so they're kept around for fear mongering but in general a nuke cant launch without a whole network of satalites locking on and tracking them from the moment their silos open and if a satalite can lock onto it, it can guide interception to it
>>108261532simple asthey’re gonna get gavin continue skullfucking california. they’re so full of shit
>>108260633My take on this was DoW was changing terms after the fact. I see that as the more likely scenario. When you contract with the real 800lb gorilla you end up having to accept whatever they come up with, esp if they have other options. Dario should have understood this much earlier when he set up the government models, they were an early entrant. Guess he's learned that now.
>>108261591>>108259976Lol this is te kind of 800lb gorilla stuff I'm talking about. > If not, officials have said they may use government authority to force Anthropic to hand over the technology anyway — while also blacklisting the company from future defense work.
>>108259925well in a capitalist societ, the ceo's should be in charge. the government works on behalf of the corpos.
>>108261591>the contract stated the ceo can control the nukes>dow is bad
I've been watching claude push boulders in a circle for 2 weeks because it completely lacks any visual memory or forward planning, but having it in charge of nuclear warheads is fine too I guess.
>>108259925This is worse than I thought. Honestly, all the Anthropic executives should be arrested for treason.
>>108262794>thinking that the Claude you have access to is the same one Pentagon useslmao
>>108262828>falling for the meme that the military always has the better toys and is 20 years ahead of the publicanon, if anthropic actually has a model that can think, give me one reason they wouldn’t release it and absolutely fucking destroy any scrap of competition until the end of time (or until chinks reverse engineer it)
>>108259952These dudes all have massive egos especially if they're jeets.In reality it's an inferiority complex though, going all the way back to when Chad would bully them during recess and take all the girls.
>>108259925anthro ceo is an autistic jew who thought he could leverage an extra trillion dollars for a scenario that will never happen. Typical of every "founder" CEO type, he can't form a sentence extemporaneously and he's just as psy-opped as the rest of the monkeys. Just another psychotic savant.
>>108262796>treason is when you want the government to either respect the contract they signed or just cancel it like a normal person
>>108259925> Nuke inbound?> Better hand this interception job off to a company who has no experience doing that and not the actual military who has prepared for that exact scenario for the last 70 years and has the equipment to do so.Do these retards really?
>>108263398Yes, when it comes to weapons of mass destruction.
>>108260613they're all glowiesthis is standard glowie showthey make the alternative and the alternative to the alternative
>>108260613anthropic is partnered with palantir and has been for yearsi think their stance on safety is probably better than others', but the dick sucking that's happening over this decision is ridiculous. in my opinion this decision has a lot to do with marketing. and it was an insanely good marketing move, don't get me wrong. anthropic is unfortunately making all the right moves.
>>108263415>You're right I should have used a laser based approached instead of a kinetic kill vehicle. That is absolutely on me. I'm sorry Chiraq got nuked by North Korea.
>>108263415military isn't responsible for war nor defense. a company is better equip for such matters. military is post to obey the civilian sector, which means companies, ceo's, public, etc
>>108259925I think he ment that they could call creating about an autonomous defensive system. Not that he wanted to decide over day to day operation.
>>108262796>all the Anthropic executives should be arrested for treason.And this precise attitude is why all the other AI companies should offshore. If they value their product.In fact. To think this will stop at AI is very unwise.
>>108259946We're speedrunning skynet bro
>>108260780thats true, they should set his house on fire, kill his dog and force him to share the mower with a gun to the head
what a control freak retardnow his company is fuckedeat shit safetyfaggot
>>108259925i wonder what kind of other models besides llms they develop if they could do that or it's just a hypothetical that they'd have to develop first
That fucker is Mossad for sure
>>108265637>they should set his house on fire, kill his dog and force him to share the mower with a gun to the headMurrica!
>>108259946Senile boomers run the world.Then along came the psychopathic salesmen who convinced them that they need the Latest New Thing™ because CURRENT_ENEMY also have it and they're gonna destroy us if we don't match and beat them.
>Hey we want to use your chatbot to write emails >"Well did you also know we think it could become a sentient fully-automated super death machine within like 1 year, trust me bro">Ok can we use it for that then>"Lol no">Cool we're going to blacklist your company from the economy Jew larpers owned
>>108259946>>108265716It's more along the lines of having it develop other AI like e.g. doing optimizations on computer vision algorithms or processing data to determine targets. I could see some rare cases where you'd want reasoning LLMs to help make a tactical decision though, but on the order of minutes not seconds
>>108259925It's another embarrassing and stupid whataboutism that displays how profoundly little the administration knows about literally everythingYou would not use claude to intercept a ballistic missile heading towards the US. You wouldn't use any LLM to do this, because speed and timing are so critical. Are you going to let the AI sit there for minutes thinking before it tells you "sorry I can't do that" or "actually, this doesn't appear to be a missile"? Or are you going to use one of the many, viable, working systems we have to detect and intercept missiles with subsecond timing instead?
>>108265802DoD had no issue developing these things prior, and were already using actual AI before meme learning came along.Machine """Learning""" is just a giant FAGMANN scam to get infinite government funding
>>108265817The main purpose of AI is to replace programmers, they can save a lot of money getting rid of them if they can get LLMs to write their code
>>108265804See>>108265837
it's so funny seeing leftards trying to hide their TDS by saying "this administration" like everybody couldn't see through it. it really puts in perspective how simple minded they are when they think such basic tricks can actually work with real human beings
>>108259925Instead we'll now get>You're right, I completely missed the nuke and instead nuked ourselves as well.Thanks openAi!
>>108261366Well then you beat the shit out of then.
Dario didn't like how Trump admin used his AI model to raid and capture Maduro.
>>108259925Should I be cancelling my ChatGPT subscription? I don't want to support them if they're going to use the tech to spy on Americans
>>108259925>AI IS THE FUTURE>DA FUTURE GOY>HOLY FUCKING SHIT IT DOES EVERYTHING>uhm a meat space person has to uhm call another meat space person to do that. What?
>>108259986It was an interview question, and he answered wrong.
>>108262796They're doing the right thing by trying to stop some 105 IQ morons in the DoW from letting a literal chatbot make world ending decisions.
>>108267312>Well then you beat the shit out of then.Sure thing.Second it reaches for my cookies.Up until then .... This is your problem chump. When it becomes my problem you're not going to like how I solve it.>>108267341>I don't want to support themYes. That's what you will be doing if you pay them.>>108267630>DoW from letting a literal chatbot make world ending decisions.Evidence that's the case?I'm hearing a lotta bullshit banded about that doesn't make any sense: Waa nuke intercept.Okies, on one hand you can deploy a tw@bot that'll take 5min of "thinking" just to hallucenate the trajectory mathORThere's the mature systems already in place like spacefence, which no part of claude could feasibly replicate or enhance.All forwarded arguments are of such bent. Realistical applications for this technology would be running mass spam campaigns to drown out any dissent with the "right" opinions. Spotting patterns in masses of data that then hand off to someone who decides what to do about that pattern is about it's peak potential.
>>108259925>anthropig
>>108264989>Would you like me to create a plan for rescue of trapped civilians and distribution of medical aid?
Uh oh! Stiiiinkyyyyyy!!!
