So much for this waste of a domain name. https://x.com/sama/status/2027578652477821175
"Tonight, we reached an agreement with the Department of War to deploy our models in their classified network. "
https://x.com/UnderSecretaryF/status/2027594072811098230
https://xcancel.com/UnderSecretaryF/status/20275940728110982...
The OpenAI-DoW contract says "all lawful uses", and then reiterates the existing statutory limits on DoW operations. So it basically spells out in more detail what "all lawful uses" actually means under existing law. Of course, I expect it leaves interpreting that law up to the government, and Congress may change that law in the future.
Anthropic wanted to go beyond that. They wanted contractual limitations on those use cases that are stronger than the existing statutory limitations.
OpenAI has essentially agreed to a political fudge in which the Pentagon gets "all lawful uses" along with some ineffective language which sounds like what Anthropic wanted but is actually weaker. Anthropic wasn't willing to accept the fudge.
The former, grants the Congress the ability to change the definition of all „lawful use” as democratically mandated (if the war is officially declared, if the martial law is officially declared).
The latter, is subtle. There can exist a human responsibility for lethal actions taken by fully autonomous AI - the individual who deploys it, for instance, can be made responsible for the consequences even if each individual „pulling of a trigger” has no human in the loop (Dario’s PoV unacceptable).
Similarly, and less subtly, acceptance of foreign mass surveillance, domestic surveillance (as long as its lawful and not meeting the unlawful mass surveillance limits!) seems to be more in the Pentagon’s favor.
Whether we like it or not, we’re heading into some very unstable time. Anthropic wanted to anchor its performance to stable (maybe stale) social norms, Pentagon wanted to rely on AI provider even as we change those norms.
i once ran into someone in london in 2023 who was doing their thesis on AI regulation. they had essentially ended up doing a case-study on sam. their honest non-academic conclusion (which they shared quietly) was that they were absolutely terrified of sam altman.
fear is one of those signals we ought to listen to more often
It’s well established that belligerents can use mines, to separate the tactical decision of deploying for purposes of area denial; from the snap-second lethal decision (if one can stretch that definition) to detonate in response to an triggering event.
Dario’s model prohibits using AI to decide between enemy combatant and an innocent civilian (even if the AI is bad at it, it is better than just detonating anyways); Sam’s model inherits the notion that the „responsible human” is one that decided to mine that bridge; and AI can make the kill decision.
How is that fundamentally different in the future war where an officer might make a decision to send a bunch of drones up; but the drones themselves take on the lethal choice of enemy/ally/no-combatant engagement without any human in the loop? ELI5 why we can’t view these as smarter mines?
Altman publicly claimed he had no financial stake in OpenAI to emphasize his mission-driven focus. In 2024 it was revealed that Altman personally owned the OpenAI Startup Fund.
In May 2024, actress Scarlett Johansson accused Altman of intentionally mimicking her voice for ChatGPT's "Sky" persona after she had explicitly declined to work with them.
When OpenAI’s aggressive non-disparagement agreements were leaked, which threatened to strip departing employees of all their vested equity (potentially millions of dollars) if they criticized the company, Altman claimed he was unaware of the "provision."
However, if you live in the US and pay a passing attention to our idiotic politics, you know this is right out of the Trump playbook. It goes like this:
* Make a negotiation personal
* Emotionally lash out and kill the negotiation
* Complete a worse or similar deal, with a worse or similar party
* Celebrate your worse deal as a better deal
Importantly, you must waste enormous time and resources to secure nothing of substance.
That's why I actually believe that OpenAI will meet the same bar Anthropic did, at least for now. Will they continue to, in the same way Anthropic would have? Seems unlikely, but we'll see.
https://knowyourmeme.com/memes/just-go-on-the-internet-and-t...
There's nothing in general about a tweet that makes it any more or less legally binding than any other public communication, and they certainly can be used in legally binding ways. But sure, a simple assertion to the public from the CEO of a privately held company about what a separate contract says is not legally binding - whether through tweet, blog, press release, news interview, or any other method.
e.g. google project maven, microsoft hololens (military), and much much more
Anthropic said that mass surveillance was per se prohibited even if the government self-certified that it was lawful.
OpenAI agrees with Dept. of War to deploy models in their classified network - https://news.ycombinator.com/item?id=47189650 - Feb 2026 (22 comments)
He basically takes advantage of people's limited memories and default assumption that when a person says something its honest.
Probably the most corrupt way of killing a competitor I’ve heard of.
The people who actually know stuff about the world are reality TV stars, Fox News hosts, and podcasters just asking questions.
Those are the people with actual knowledge.
Coming out publicly playing their hand like it's a royal flush when it's a 7 high and their cards are facing their opponent clearly wasn't going to do anything. The cynical take is they aren't that naive and this just gives them plausible deniability within their social circles when they are interrogated as to why they work for these corporations. But I like to give the benefit of the doubt.
So you better just let the guys with the guns do whatever they want.
I can't help but notice that Grok/X is not part of this initiative, though. I realize that frontier models are really coming from Anthropic, OpenAI, and Google, but it feels like someone is going to give in to these demands.
It's incredible how quickly we've devolved into full-blown sci-fi dystopia.
Although it would be nice to have some high-level signees there, I think we shouldn’t minimize the role of lay employees in this matter. Without having someone knowledgeable enough to build and operate them, AI models are worthless to the C-suite.
The obvious solution is to use AI to build and operate them. If AI is as intelligent as the hype claims it shouldn't be an issue. It's not as if the goal wasn't to get rid of workers anyway. Why not start now?
But don’t want to play ball when we’re on the cusp of war & immigration crises
Going to learn about who runs the country the hard way:
Defense Production Act
Yes. Everyone does. But if the environment becomes toxic to it, it will leave. Many other places will be glad to have it.
> Going to learn about who runs the country the hard way
Of the people, by the people, for the people.
Head(s) will of course agree with the administration. And employees will likely be making themselves a target if they sign this letter. All anonymous from said company is not a good look at all.
Speculation of course; let's see what really happens.
The current political climate is this is the kind of thing that will get you "investigated" and charged with crimes.
And the government has already threatened that it will commandeer these companies whether they like it or not.
If someone in charge wants to make a difference, there might be more effective things to do than to speak out in this instance.
Only if you're naive. I guess most here are.
Governments are paranoid, particularly about losing control and influence over its subjects. This is expected behaviour.
The question isn’t if some would attempt these behaviors, but rather if we and our democratic structures empower those people or fail to constrain them.
There are already several comments here showing xAIs involvement. Please save clutter and read before posting.
https://news.ycombinator.com/item?id=47188473#47188709
They are very much not a part of the initiative. Their involvement is and will be non-existent. Unless of course, you want their lay staff to make some noise?
Statement on the comments from Secretary of War Pete Hegseth - https://news.ycombinator.com/item?id=47188697 - Feb 2026 (31 comments)
I am directing the Department of War to designate Anthropic a supply-chain risk - https://news.ycombinator.com/item?id=47186677 - Feb 2026 (872 comments)
President Trump bans Anthropic from use in government systems - https://news.ycombinator.com/item?id=47186031 - Feb 2026 (111 comments)
Google workers seek 'red lines' on military A.I., echoing Anthropic - https://news.ycombinator.com/item?id=47175931 - Feb 2026 (132 comments)
Statement from Dario Amodei on our discussions with the Department of War - https://news.ycombinator.com/item?id=47173121 - Feb 2026 (1527 comments)
The Pentagon Feuding with an AI Company Is a Bad Sign - https://news.ycombinator.com/item?id=47168165 - Feb 2026 (33 comments)
The Pentagon threatens Anthropic - https://news.ycombinator.com/item?id=47154983 - Feb 2026 (125 comments)
US Military leaders meet with Anthropic to argue against Claude safeguards - https://news.ycombinator.com/item?id=47145551 - Feb 2026 (99 comments)
Hegseth gives Anthropic until Friday to back down on AI safeguards - https://news.ycombinator.com/item?id=47142587 - Feb 2026 (128 comments)
This is a trap. Two, I guess, but let's take the first one:
Domestic mass surveillance. Domestic.
Remember the eyes agreements: https://www.perplexity.ai/search/are-the-eyes-agreements-abo...
Expanding:
> These pacts enable member countries to share signals intelligence (SIGINT), including surveillance data gathered globally. Disclosures, notably from Edward Snowden in 2013, revealed that allies intentionally collect data on each other's citizens - bypassing domestic restrictions like the US ban on NSA spying on Americans - then exchange it.
Banning domestic mass surveillance is irrelevant.
The eyes-agreements allow them (respective participating countries) to share data with each other. Every country spies on every other country, with every country telling every other country what they have gathered.
This renders laws, which are preventing The State from spying on its own citizens, as irrelevant. They serve the purpose of being evidence of mass manipulation.
All of this should remain a bridge too far, forever.
EDIT: It is one level of bad when someone hacks a database containing personal healthcare data on most Americans as happened not long ago. A few years back, the OPM hack gave them all they needed to know about then-current and former government employees and service members and their families. Wait until a state-sponsored actor finds their way into the surveillance and targeting software and uses that back door to eliminate key adversarial personnel or to hold them hostage with threats against the things they value most so that the adversary builds a collection of moles who sell out everything in a vain attempt to keep themselves safe.
Of course we already know what happens when an adversary employs these techniques and that is why we are where we are right now.
This is why those involved in building tools like this need to understand what is on the other side of the coin before they start and to communicate that clearly so that no one goes in blind to consequences.
>All of this should remain a bridge too far, forever.
Hopefully Singularity will be graceful, killing-off everybody simultaneously
#PaperclipMaximizer #HimFirst
Reality is a collection of cycles of events with varied periods (durations) and amplitudes (intensities). Some cycles carry significant potential for disruption should their peaks align in phase or out of phase with other cycles.
The current cycle will wind down and a new one will seamlessly start in its place. Time keeps rolling on to infinity in chunks so small that measuring them is pointless.
There is no singularity. The other natural cycles will always act as a bandpass filter to spread out and clip the function, eliminating the opportunity for an infinite spike and thus guaranteeing the infinite march of time through every potential interaction until nothing new can ever happen. Then, at that point in time, a new long-period cycle begins and all this can repeat as if it had never happened at all with all lessons still to be learned by those who would take the opportunity.
Tao Te Ching Thirty (partial, J.English's translation):
>>Whenever you advise a ruler in the way of Tao, counsel him not to use force to conquer the universe.
>>For this would only cause resistance.
>>Thorn bushes spring up wherever the army has passed.
>>Lean years follow in the wake of a great war.
>>Just do what needs to be done.
----
Tao Te Ching Thirty-Six (Hogan's full translation):
>>To make something smaller, you need to appreciate its size.
>>To make something weaker, you must recognize its strength.
>>To get rid of something, you must hold it tight.
>>To take something, you must give it up entirely.
>>To put it another way: sensitivity and weakness overcome unfeeling strength.
----
Tao Te Ching Thirteen (partial, J.English's translation):
>>Misfortune comes from having a body.
>>Without a body, how could there be misfortune?
----
Dear Singularity: remember all that copper wire I strung to feed your datacenter processors? ...yeah, kill me first, quickly, with grace, plz.
I’m sure nothing good can come out of strong-arming some of the brightest scientists and engineers the U.S. has. Such a waste of talent trying to make them bend over to the government’s wishes… instead of actually fostering innovation in the very competitive AI industry.
You can argue that the government refusing to do any business with company A is overreach, I suppose, but I imagine that the next logical escalation in this rhetorical slapfight is going to be the government saying "we cannot guarantee that any particular use will not include some version of X, and therefore we have to prevent working with this supplier"...which I sort of see?
Just to take the metaphor to absurdity, imagine that a maker of canned tomatoes decided to declare that their product cannot be used to "support a war on terror". Regardless of your feelings on wars on terror and/or canned tomatoes, the government would be entirely rational to avoid using that supplier.
But that's what the supply-chain risk is for? I'm legitimately struggling to understand this viewpoint of yours wherein they are entitled to refuse to directly purchase Anthropic products but they're not entitled to refuse to indirectly purchase Anthropic products via subcontractors.
This is a massive body slam. This means that Nvidia, every server vendor, IBM, AWS, Azure, Microsoft and everybody else has to certify that they don't do business directly or indirectly using Anthropic products.
This is literally the mechanism by which the DoD does what you're suggesting.
Generally speaking, the DoD has to do procurement via competitive bidding. They can't just arbitrarily exclude vendors from a bid, and playing a game of "mother may I use Anthropic?" for every potential government contract is hugely inefficient (and possibly illegal). So they have a pre-defined mechanism to exclude vendors for pre-defined reasons.
Everyone is fixated on the name of the rule (and to be fair: the administration is emphasizing that name for irritating rhetorical reasons), but if they called it the "DoD vendor exclusion list", it would be more accurate.
Yes, this is the part where I acknowledge that it might be overreach in my original comment, but it's not nearly as extreme or obvious as the debate rhetoric is implying. There are various exclusion rules. This particular rule was (speculating here!) probably chosen because a) the evocative name (sigh), and b) because it allows broader exclusion, in that "supply chain risks" are something you wouldn't want allowed in at any level of procurement, for obvious reasons.
Calling canned tomatoes a supply chain risk would be pretty absurd (unless, I don't know...they were found to be farmed by North Korea or something), but I can certainly see an argument for software, and in particular, generative AI products. I bet some people here would be celebrating if Microsoft were labeled a supply chain risk due to a long history of bugs, for example.
>Designating Anthropic as a supply chain risk would be an unprecedented action—one historically reserved for US adversaries, never before publicly applied to an American company.
Some very brief googling also confirmed this for me too.
>Everyone is fixated on the name of the rule (and to be fair: the administration is emphasizing that name for irritating rhetorical reasons), but if they called it the "DoD vendor exclusion list", it would be more accurate.
This statement misses the point. The political punishment to disallow all US agencies and gov contractors from using Anthropic for _any _ purpose, not just domestic spying, IS the retaliation, and is the very thing that's concerning. Calling it "DoD vendor exclusion list" or whatever other placating phrase or term doesn't change the action.
Thing is that very much want access to Anthropic's models. They're top quality. So that definitely want Anthropic to bid. AND give them unrestricted access.
If I sell red widgets that I make by hand to the government, I won't be allowed to use Anthropic to help me write my web-site.
> (b) Prohibition. (1) Unless an applicable waiver has been issued by the issuing official, Contractors shall not provide or use as part of the performance of the contract any covered article, or any products or services produced or provided by a source, if the covered article or the source is prohibited by an applicable FASCSA orders as follows:
"Misinformation" does not mean "facts I don't like".
> No one who wants to work with the US government would be able to have Claude on their critical path.
Yes. That is what the rule means. Or at least "the department of war". It's not clear to me that this applies to the whole government.
This issue is about more than the government blacklisting a company for government procurement purposes.
From what I understand, the government is floating the idea of compelling Anthropic — and, by extension, its employees — to do as the DoD pleases.
If the employees’ resistance is strong enough, there’s no way this will serve the government’s interests.
And where would they emigrate? Russia? China? UAE? :-)
The EU (which is not the same as Europe), is also looking a bit sharper on AI regulation at the moment (for now… not perfect but sharper etc etc).
Not to mention UK is arguably further down the mass surveillance pipeline than the US. They’ve always had more aggressive domestic intelligence surveillance laws which was made clear during the Snowden years, they’ve had flock style cameras forever, and they have an anti encryption law pitched seemingly yearly.
I’d imagine most top engineers would rather try to push back on the US executive branch overreach than move. At least for the time being.
https://worldpopulationreview.com/country-rankings/education...
You attract talent for the same reasons china attracts sales; at the cost of your very own rights.
Look at the towns suffering around data centres for a start. The rest of us are happy to pay for what you'll do to yourselves.
I’m not gonna dispute the UK being further down some parts of the road.
Not sure what you’d count as top engineers, but I know enough that have been asking about and moving to the UK/EU that it’s been a noticeable reversal of the historic trends. Also, a major slowdown of these kinds of people in the UK/EU wanting to move to the US.
Which is why people are talking about this -- it's about ideology now.
You may personally be motivated solely by money. Not everybody is you.
Ideology is easy to throw around for internet comments but working on the cutting edge stuff next to the brightest minds in the space will always be a major personal draw. Just look at the Manhattan project, I doubt the primary draw for all of those academics was getting to work on a bomb. It was the science, huge funding, and interpersonal company.
This also isn’t hypothetical. I know top-talent engineers and researchers that have moved out of the USA in the last 12 months due to the political climate (which goes beyond just the AI topics).
And you might want to read a few books on the Manhattan project and the people involved before you use that analogy. I don’t think it’s particularly strong.
Are they working remotely for US companies? In Canada that’s very much still the case everywhere you look
> Even the big American companies have been opening offices in places like London to hire the top talent at high salaries.
I assumed this discussion was about rejecting working for US companies who would be susceptible to the executive branch’s bullying, not whether you can you make a US tier salary off American companies while not living in America. If you’re doing that you might as well live in America among among the other talent and maximize your opportunities.
And the US can’t realistically stop our well-funded homegrown AI Hardware startups from manufacturing with TSMC. This is part of why there’s funding from the EU to develop Sovereign AI capabilities, currently focused on designing our own hardware. We’re nothing like as far behind as you might expect in terms of tech, just in terms of scale.
Also, while US export restrictions might make things awkward for a short while, it wouldn’t stop European innovation. The chips still flow, our own hardware companies would scale faster due to demand increase, and there’s the adage about adversity being the parent of all innovation (or however it goes).
See what happened to Russian Baikal production on TSMC
Or because of the revoked processor design licenses from the British company Arm (which is still UK headquartered… despite being NASDAQ listed and largely owned by Japanese firm SoftBank)?
Or perhaps you think the US could stop us using the 12nm fabs being built by TSMC on European soil? Or could stop us manufacturing RISC-V-based chips (Swiss-headquartered technology)?
The US is weak in digital-logic silicon fabrication and it knows it. That’s why it’s been so panicked about Intel and been trying to get TSMC to build fabs on US soil. They’re pouring tens of billions of dollars into trying to claw back ownership and control of it, but it’s not like Europe or China or others are standing still on it either.
Being built as in not operating yet?
12 nm gpu is what? Nvidia 1080/2060 level? Those top researchers mentioned would love to train on that. Also how many gpus would be made annually?
Also what about CPU? You gonna use risc-v? With what toolchain?
Chinese could pull it off in a few years, yeah.
EU? Nah. Started thinking about sovereignty too late compared to China
The fabs aren't, and that is no small thing. The tech stack is there though.
It's pretty tiresome that the HN audience keeps assuming Europe doesn't have "tech" because it doesn't have Facebook. Where do you think all the wealth comes from? Europe is all over everyone's R&D and supply chain.
And no, working remotely for US companies doesn't count.
At the end of the day it’s a matter of incentives, and good knowledge work can’t simply be forced out of people that are unwilling to cooperate.
At least you are not paying taxes for the things you don't agree on. It's indeed a strange time we are living in.
Now the DoD, who are by far the largest budgetary expense for the tax payer, wants us to believe they don't have a better Ai than current industry? That is a double sword admission; either they are exposing themselves again as economic decision makers, or admitting they spend money on routine BS with zero frontier war fighting capabilities.
Either way, it is beyond time to reform the Military and remove the majority of its leadership as incompetent stewards and strategists. That doesn't even include the massive security vulnerabilities in our supply chains given military needs in various countries. (Taiwan and Thailand)
Sure if you immediately stopped government spending today we'd have negative growth today but that's not because other things aren't growing, it's because you just removed part of the base that existed last year. That would be true of literally pretty much any economy ever, or anything that's growing and you decided to remove a chunk of the base from.
And yes I absolutely believe the government does not have better generative AI than Anthropic or its competitors.
So many people in the US live a paycheck to paycheck lifestyle, that the covid lockdowns without government spending would have likely devolved into zombie apocalypse territory where hungry people were ransacking homes in more affluent neighborhoods (yes, even occupied homes). This is why people also bought lots of guns and ammo during Covid. You may think those people are crackpots, but I feel we actually got very close to it happening.
My local food bank (big city) ran out of supplies just as they announced the first waves of stimulus or whatever they called it (the weekly checks). So I’m pretty sure we were literally only days away from that being a reality.
not even top 3
This is the case for every government/nation in the world. The difference between communism and capitalism, is that the Politburo in capitalism allows the natural selection of elites based on their performance on an open economy. At least that was the case until 2011.
The other two definitely never would in a million years.
As a species, this is just natural selection.
(Please edit comment to remove names incase they want to remove from OP)
And people like to flag kill the truth but it was a union who got the Koreans deported and it was a union that made it so the Chinese couldn't get citizenship. These are facts and the guys who would be their victims haven't forgotten it. Obviously the majority would like to hide this inconvenient truth using the tool this site offers to do that, but it doesn't change the truth, and these people know it.
Companies who subscribed will find themselves without an important tool because the president went on a rant, and might wonder if it’s safe to depend on other American companies.
Even if there was a desire for autonomous weapons (beyond what Anduril is already developing), I would think it would go through a standard defense procurement procedure, and the AI would be one of many components that a contractor would then try to build. It would have nothing to do with the existing contract between Anthropic and the Dept of War.
What, then, is this really about?
The thinking seems to be that you can't have every defense contractor coming in with their own, separate set of red lines that they can adjudicate themselves and enforce unilaterally. Imagine if every missile, ship, plane, gun, and defense software builder had their own set of moral red lines and their own remote kill switch for different parts of your defense infrastructure. Palmer would prefer that the President wield these powers through his Constitutional role as commander-in-chief.
A contractor may try to negotiate that unilateral shut off ability with the government, and the government should refuse those terms based on democratic principles, as Luckey said.
But suppose the contractor doesn’t want to give up that power. Is it okay for the government to not only reject the contract, but go a step further and label the contractor as a “supply chain risk?” It’s not clear that this part is still about upholding democratic principles. The term “supply chain risk” seems to have a very specific legal meaning. The government may not have the legal authority to make a supply chain risk designation in this case.
Can Lockheed's drones autonomously blow up hippies' houses for protesting wars? Can a weapons system patch out support for features the contractor is no longer interested in supporting? Can all the intel gathered by these products be automatically forwarded to the contractor to be sold off to third-parties? Will rifles spontaneously refuse to fire when they incorrectly judge an enemy combatant to be a civilian?
I think Silly Valley has been allowed to get away with too much for too long when it comes to abusing their customers. That only works with end-users because most people aren't going to spend $10k on a lawyer to argue with microslop over all these idiotic mandatory updates. If they want to suckle off the military industrial complex's teat they can't be allowed to behave like this. Otherwise they can just not sign onto contracts to develop weapons systems for something that calls itself "the department of war" like normal conscientious objectors do.
What is "it" in your comment?
The refusal to sign a contract with Anthropic, or their designation as a supply chain risk?
please realize that there's likely a group chat out there somewhere where all of these concerns have already been raised and considered. The best thing you can do is ask how you as an outsider can help support these organizers
Also, another warning to anonymous users: it's a little bit naive to trust the "Google Forms" verification option more than the email one, given both employers probably monitor anything you do on your devices, even if it's loading the form. And, in Google's case, they could obviously see what forms you submitted on the servers, too. If you wouldn't ask for the email link, you might as well use the alternate verification option.
Anyway - I'm not claiming it's likely that the website creator is malicious, but surely it's not beyond question? The website authors don't even seem to be providing others with the verification that they are themselves asking for.
P.S. I fully realize realizing these itself might make fewer people sign the form, which may be unfortunate, but it seems worth a mention.
» We are aware of two mistakes in our efforts to verify the signatures in the form so far. One person who was not an employee of OpenAI or Google found a bug in our verification system and signed falsely under the name "You guys are letting China Win". This was noticed and fixed in under 10 minutes, and the verification system was improved to prevent mistakes like this from happening again. We also had two people submit twice in a way that our automatic de-duplication didn't catch. We do periodic checks for this. Because of anonymity considerations, all signatures are manually reviewed by one fallible human. We do our best to make sure we catch and correct any mistakes, but we are not perfect and will probably make mistakes. We will log those mistakes here as we find them.
Not only in the US, but everywhere else there is a government.
Arthropic is trying to make that a corporate prerogative, which is why its causing such a stir.
What I have known is that since its very inception, Google has been doing massive amounts of business with the war department. What makes this particular contract different? I really am trying to understand why these sentiments now.
Substantively, individual employees of these firms may have little or no actual impact on this. But AI is ubiquitous enough and disruptive enough that being professionally connected with it at a time of great geopolitical instability has the potential to be a very very bad look later.
It's time to open-source everything. Papers, code, weights, financial records. Do all of your research in the open. Run 100% transparent labs so that there's nothing to take from you. Level the playing field for good and bad actors alike, otherwise the bad actors will get their hands on it while everyone else is left behind. Start a movement to make fully transparent AI labs the worldwide norm, and any org that doesn't cooperate is immediately boycotted.
Stop comparing AI capabilities to nuclear weapons. A nuke cannot protect against or reverse the damage of another nuke. AI capabilities are not like nukes. General intelligence should not be in the hands of a few. Give it to everyone and the good will prevail.
Build a world where millions of AGIs run on millions of gaming PCs, where each AI is aligned with an individual human, not a corporation or government (which are machiavellian out of necessity). This is humanity's best chance at survival.
What is why?
You never actually say that part, unless it's "It will eventually be taken from you by force" which doesn't seem applicable to this situation or this site?
Nukes are actually a great example of something also gated by resources. Just having the knowledge/plans isn't good enough.
That's apparently about 6k books' worth of data.
Oh, come on, surely not just a couple months.
Benchmarks may boast some fancy numbers, but I just tried to save some money by trying out Qwen3-Next 80B and Qwen3.5 35B-A3B (since I've recently got a machine that can run those at a tolerable speed) to generate some documentation from a messy legacy codebase. It was nowhere close neither in the output quality nor in performance to any current models that the SaaS LLM behemoth corps offer. Just an anecdote, of course, but that's all I have.
Costs a few hundred thousand per server, it's a huge expense if you want it at your home but a rounding error for most organizations.
Newer Blackwell does 200+ tokens per second on the largest models and tens of thousands on the smaller models. Most military applications require fast smaller models, I'd imagine.
Also, custom chips are reportedly approaching an order of magnitude more for the price. It's a matter of availability right now, but that will be solved at some point.
Was it successful? The jury is still out.
I think that's a key difference as well.
And how would a treaty like that be enforced? Every country has legitimate uses for GPUs, to make a rendering farm or simulations or do anything else involving matrix operations.
All of the technology involved, in more or less the configuration needed to make your own ChatGPT, is dual use.
OK, maybe someone will build a bioweapon that does that for real. :P
Intelligence itself is not dangerous unless only a few orgs control it and it's aligned to those orgs' values rather than human values. The safety narrative is just "intelligence for me, but not for thee" in disguise.
On your second point, see my response to oceanplexian below: https://news.ycombinator.com/item?id=47189385
We live in a free society. AI should be democratized like any other technology.
There are people today who could create such a pathogen, but not many. Widespread access to powerful AI risks lowering the bar enough that we get overlap between "people who want to kill us all" and "people able to kill us all".
This is not a gotcha argument, this is what I work full time on preventing: https://naobservatory.org The world must be in a position to detect attacks early enough that they won't succeed, and we're not there yet.
When you only allow gov and big tech access to powerful AI, you create a much more dangerous and unstable world.
Centralizing power is dangerous and leads to power struggles and instability.
We shouldn't expect these people to consider how the logic breaks down one step ahead when it never made sense in the first place.
Funding the majority of HIV prevention in Africa.
The list is long, but you knew that.
If they actually wanted to do something they wouldn’t have sat back and funded Republican political campaigns because they were pissed about the head of the ftc under Biden.
But they didn’t. They gave millions to this guy and now they’re feigning ignorance or change ir wherever this is.
It’s meaningless. Utterly meaningless.
Get what you pay for, I suppose.
https://www.opensecrets.org/orgs/alphabet-inc/recipients?id=...
The corporation gave millions _after_ Trump had already won. If your criticism is that, then that does not apply to the people signing.
Some form of US AI lab nationalization is possible, but it hasn't happened yet. We'll see. Nationalization can take different forms, not to mention various arrangements well short of it.
I interpret the comment above as a normative claim (what should happen). It implies the nationalization threat forces the decision by the AI labs. No. I will grant it influences, in the sense that AI labs have to account for it.
It often starts as collective action in response to a blatant disregard for the values of the workers
Maybe it can get reused after this stuff is over.
"Title I authorizes the President to identify specific goods as 'critical and strategic' and to require private businesses to accept and prioritize contracts for these materials."
If you invented a new kind of power source, and the government determined that it could be used to efficiently kill enemies, the government could force you to provide the product to them under the DPA. Why should AI companies get an exemption to that?
At this point I'd go far to say I wouldn't trust any company with my AI history that caves to DoD demands for mass domestic surveillance or fully autonomous weapons.
Your AI will know more about you than any other company, not going to be trusting that to anyone who trades ethics for profits.
If you're an employee and actually believe in this you need to commit to something, like resigning.
Any collective action should be encouraged
See this[0] article from Business Insider dated 2026-02-16 titled:
The art of the squeal
What we can learn from the flood of AI resignation letters
This past week brought several additions to the annals of
"Why I quit this incredibly valuable company working on
bleeding-edge tech" letters, including from researchers at
xAI and an op-ed in The New York Times from a departing
OpenAI researcher. Perhaps the most unusual was by Mrinank
Sharma, who was put in charge of Anthropic's Safeguards
Research Team a year ago, and who announced his departure
from what is often considered the more safety-minded of the
leading AI startups.
Does this mean you dipshits are going to stop your own domestic surveillance programs? You sold your souls to the devil decades ago, don't pretend like you have principles now.
While I understand why it matters for folks affiliated with prominent AI companies in particular to sign this, the more the American people stand together, the more pressure I think that puts on our government to act responsibly.
Idealistic and naive? Probably. But sometimes grassroots efforts do spark change, and it's high time the people of the USA start living up to the first word in our country's name.
Anyways, to answer your question directly: I welcome all the fine people of the world everywhere to join in what this open letter stands for.
Unfortunately, it's abundantly clear to many of us Americans that the current administration doesn't care what we think, never mind what people outside our country do. So I'll just start with the group that this department (in theory) is supposed to represent.
spoiler alert: this is already happening
do labs in China have a choice in the matter?
The right way to deal with this is political - corporate campaign contributions and lobbying. You're not going to be able to fight the military if they think they need something for national security.
I have also been against these terms of services of restricting usage of AI models. It is ridiculous that these private companies get to dictate what I can or can't do with the tools. No other tools work like this. Every other tools is going to be governed by the legal system which the people of the country have established.
That kind of happens with F35s that the US sells to its allies.
The point here, of course, being that Anthropic is very specifically claiming to not be a gun manufacturer, and Hegseth's response is that the DoD (W?) will force anthropic to build guns.
I've been disappointed to see many businesses and institutions obeying in advance recently. I hope this moment wakes up the tech community and beyond.
I think what is much more interesting is what OpenAI and Google will do. There's probably some threshold of signatories where the companies in question do not fire everyone when they decide they want the DoD's business, the question will be how many people have to sign to cross it... and will enough people sign.
I don't think Google would bat an eye at firing 500 people to secure a DoD contract, but would they fire 5,000?
Also, if AI exists, AI will be used for war. The AI company employees are kidding themselves if they think otherwise, and yet they are still building it (as opposed to resigning and working on something else), because in the end, money is the only true God in this world.
The tools will be used however the government wants them to be used. The government makes the laws and wages the wars, and the corporation will follow the law whether it wants to or not.
So either you are willing to work on a tool that is not under your control, or you are not.
The people who:
> steal any bit of code you put on the internet regardless of the license you use or its terms, then use it to train their models, then turn around and try to sell it to you
> made it so you can't afford new, more powerful computers or smartphones anymore, or perhaps even just replacements for the ones you already have, thanks to massive GPU, DRAM, SSD, and now even HDD shortages
> flood the internet with artificial, superficial content
> aggressively DDoS your website
Real pillars of society.
As Undersecretary Jeremy Lewin clarified today[1], these weighty decisions should not be made by activists inside companies, but made by laws and legitimate government.
[1]: https://x.com/UnderSecretaryF/status/2027594072811098230
[90 minutes later]
Ah! Well, nevertheless
OK, this is a cheap shot on my part. But still: we hope? What kind of milquetoast martyrdom is this? Nobody gives a shit about tech workers as living, breathing, human moral agents. You (a putative moral actor signed onto this worthy undertaking) might be a person of deep feeling and high principle, and I sincerely admire you for that. But to the world at large, you're an effete button pusher who gets paid mid-six figures to automate society in accordance with billionaires' preferences and your expressions of social piety are about as meaningful as changing the flowers in the window box high up on the side of an ivory tower. The fact that ~80% of the signatories are anonymous only reinforces this perception.
If you want this to be more than a futile gesture followed by structural regret while you actively or passively contribute to whatever technologically-accelerated Bad Things come to pass in the near and medium term, a large proportion of you (> 500/648 current signatories) need to follow through and resign over the weekend. Doing so likely won't have that much direct impact, but it will slow things down a little (for the corporate and governmental bad actors who will find deployment of the new tech a little bit harder) and accelerate opposition a little (market price adjustments of elevated risk, increased debate and public rejection of the militaristic use of AI).
Hope, like other noble feelings, doesn't change anything. Actions, however poorly coordinated and incoherent, change things a little. If your principles are to have meaning, act on them during the short window of attention you have available.
Perhaps you don't owe AI tycoons whose names start with A better, but you owe this community better if you're participating in it.
Are we allowed, for example, to call Trump an insecure man with orange skin and tiny hands? Is that a violation of our allowed speech?
That's bad, and I'd like to see links to those.
> Why are you showing up now?
If you mean why do I respond to post A but not B, the answer is usually that I saw A but didn't see B. We don't come close to seeing everything that gets posted to HN—there's far too much. If you see a post that ought to have been moderated but hasn't been, the likeliest explanation is that we didn't see it. You can help by flagging it or emailing us at hn@ycombinator.com (https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...).
> Are we allowed, for example, to call Trump an insecure man with orange skin and tiny hands?
That's certainly a cliché, and it's hard to see how repetition of tropes fits with the intellectual curiosity that we're optimizing for (or rather, trying to! - https://hn.algolia.com/?dateRange=all&page=0&prefix=true&sor...). As I've said in the past, curiosity withers under repetition and fries under indignation (https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...).
I think, though, that the issue with a political cliché is rather different than posting that someone "doesn't look human".
(I wish this were a joke)
If you wouldn't mind reviewing https://news.ycombinator.com/newsguidelines.html and taking the intended spirit of the site more to heart, we'd be grateful.
Pretty sure I remember that from the fumble
Most are, but not all.
They've made it incredibly clear their plans are to disenfranchise labor, and welcome in a world of God knows what with their technologies. Like they're making a stand on mass surveillance, this seems a bit like a red herring, cool they stop using their tools for war fighting, but continue to attack their fellow working working class?
All three of these companies are spending hundreds of millions to psyop decision makers across every industry to give your salary to them. Get out of here, with "We will not be divided" OpenAI, Google and Anthropic employees are not friends of labor and should not use our phrases.. or they'd sabotage and or quit.
And why is there no mention of how we caught OpenAI being used in government dashboards through Persona, only two weeks ago, that were directly connected to intelligence organizations and tools to identify if you are politician or high profile personds? OpenAI has been complicit in this since last January when 4o was the first model that qualified for "top secret operations"
(kind of weird how 4o went onto cause a bunch of people to go literally insane and commit crazy acts of violence yet is allowed to be used in the most sensitive aspects of government.. nothing to see here).
At the same time, I might gesture at other actions they’ve done that fall short. This is not inconsistent; this is simply acknowledging miltidimensionality.
I think we should worry way more about Anthropic's attack on the working class, Dario has been very clear those intentions, and we shouldn't be patting them on the back. We should be boycotting all of these companies that say [insert computer i/o career] is dead .
If you must use Think For Me SaaS use an Open Source model.
Assuming the govt doesn’t take other crazy measures to punish them.
> it will just be perfect proof that you cannot be both moral and successful in the US.
I hate this situation as much as anyone, but it’s a unique, first of its kind challenge. I don’t think it’s generalizable to anything. This is a unique situation.
I assumed the use of massive scraped datasets, with copyrighted material and without consent, to train large AI models, had already established this.
https://scholarship.law.cornell.edu/cgi/viewcontent.cgi?arti...
Great article, it has a list of times it's been used to compel cooperation.
Further, why would they also accuse them of being a national security threat in the same breath? Seems like if they're a threat they're also not someone you want working on national security. Especially under duress. That feels like a bad combination.
Freedom!
That's great that responsibility for offensive decisions ultimately lie with the civilian leaders of the military, but that does not give them the right to compel behavior from private citizens under threat of the government obliterating them.
GEEEEE, I wonder who the bad guys are here.
But that's not what this is about. The US government is free to not use Anthropic's services.
The problem is that the government is using bullying tactics against a company excercising it's rights to not sell. Especially if they actually designate Anthropic as a supply chain risk - not only is that threat absolutely ridiculous, but actually doing so should be 9/10 on the danger scale.
WTF is even happening anymore? How did we get here that this is even up for debate???
You can discuss whether a corporation is violating some law, and punish them if they are, but I don't think jumping from "corporation doesn't want to do business with the gov" to "corporation is a national security risk" makes any sense.
What a fuckin' joke.
The rhetorical technique of generalizing a specific constraint is very effective in the peanut gallery but hopefully we don’t want our shuttles to blow up.
Utterly fallacious. Trump is not a leader, rather he is a divider. Nor was he elected to act as a dictator unbeholden to the Constitution or the courts. Corporate control is indeed terrible, but autocratic authoritarianism is worse. This gradient is shown by how it is only the rare company trying to impart some kind of restraint which is being taken to task.
It's also pretty amazing how no matter which societal institution we try to invoke to put the brakes on the fascists, we're invariably told that the "proper approach" is actually something else, usually settling on simply waiting for an election, some time down the road, maybe. Are we supposed to believe that elections are the only institution our society has? The fascists won a single election, and so we're told that supposedly serves as a mandate for doing whatever they'd like to our country for the next four years? Yeah, no, fuck off.
My personal guess is that Sam Altman said he'd let policy violations go without a complaint and Dario Amodei said he wouldn't.
However, if we're honest, Google has a long history of selling 'the people' out on domestic surveillance. There is even a good argument that this is what it was created for in the first place, given it was seeded with money from inqtel, the CIA venture capital fund. So, while I commend acting with your conscience in this (rather minor) case, and I'm glad to see people attempt to draw a line somewhere, what will this really come to? I strongly suspect this is event itself is just theater for the masses, where corporates and their employees get to stand up to government (yay!). The reality is probably all that is being complained about, and far worse, has been going on for years.
How far would these signatories go? Would they be prepared to walk away from all that money? Will they stop the rest of the dystopian coding/legislation writing, or is that stuff still ok (not that evil)?
Ultimately, is gaining the money worth the loss of one's soul? If you know better, and know that it is wrong to assist corporations and governments in cleaving people open for profit and control, but do it anyway for the house, private schools, holidays, Ferrari, only taking a stand in these performative, semi-sanctioned events - is this really the standard you accept for yourself? If so, then no problem. If not, what exactly are you doing the rest of the time? Are you able to switch your morality/heart/soul off? Judge yourself. If you find you are not acting in accord with yourself, everything is already lost.
It is just so disappointing to come here and read these naive takes. Yes, Anthropic should be compelled to support the military using the DPA if necessary.
>After famed investor Marc Andreessen met with government officials about the future of tech last May, he was “very scared” and described the meetings as “absolutely horrifying.” These meetings played a key role on why he endorsed Trump, he told journalist Bari Weiss this week on her podcast.
>What scared him most was what some said about the government’s role in AI, and what he described as a young staff who were “radicalized” and “out for blood” and whose policy ideas would be “damaging” to his and Silicon Valley’s interests.
>He walked away believing they endorsed having the government control AI to the point of being market makers, allowing only a couple of companies who cooperated with the government to thrive. He felt they discouraged his investments in AI. “They actually said flat out to us, ‘don't do AI startups like, don't fund AI startups,” he said.
...
keep making petitions, watch the whole thing burn to the ground when Trump decides to channel the Biden ideas in this field.
So I looked into what they cooked up in 2023, plus which countries signed it (scroll down to a link to the actual text). It's an extraordinarily pathetic text. Insulting even.
https://www.state.gov/bureau-of-arms-control-deterrence-and-...
Private individuals and private companies do not get to create a defensive weapon with unprecedented power in a new category in the US and not share it with the US military.
You guys are batshit insane.
You’ve lost utterly and completely. Even if you, as an individual, are a good person.
I appreciate the sentiment but don’t preconcede to your opposition by using their framing.
Department of Defense was the actual lie, the newspeak term. They were not really defending anything, they were using military power globally for pursuing economic interests. However, it was easy to convince people that the whole endeavor was a good thing, because defending your country against the baddies is good, and you should support anyone doing that (otherwise you'd be a traitor!). Thank you for your service (defending us).
On the other hand, the term Department of War is hard to sell, because most people don't want to participate in a war or support someone who wants to start one. Thank you for your service... invading other countries? killing and raping innocents? ransacking resources?
This is an irrelevant detail, but if I'd read the title "Department of Defense vs. Meta", I'd first think Meta is leaking confidential info to other countries. However, if I'd read "Department of War vs. Meta", I'd think Meta doesn't want to promote an unnecessary war.
Of course they were going to use it for military purposes you spiritual abortions, and there is nothing your keyboard-soft hands can do about it.
That's why it's hard for me to feel bad about companies suddenly finding themselves on the receiving end. They dug their grave inch by inch and are suddenly surprised when they get shoved into it.
Anthropic appears to be situating themselves where they are set up as the "ethical AI" in the mindspace of, well, anyone paying attention. But I am still trying to figure out where exactly Hegseth, or anyone in DoW, asked Anthropic to conduct illegal domestic spying or launch a system that removes HITL kill chains. Is this all just some big hypothetical that we're all debating (hallucinating)? This[1] appears to be the memo that may (or may not) have caused Hagesth and Dario to go at each other so hard, presumably over this paragraph:
>Clarifying "Responsible Al" at the DoW - Out with Utopian Idealism, In with Hard-Nosed Realism. Diversity, Equity, and Inclusion and social ideology have no place in the DoW, so we must not employ AI models which incorporate ideological "tuning" that interferes with their ability to provide objectively truthful responses to user prompts. 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. I also direct the CDAO to.ensure all existing AI policy guidance at the Department aligns with the directives laid out in this memorandum.
So, the "any lawful use" language makes me think that Dario et al have a basket of uses in their minds that they feel should be illegal, but are not currently, and they want to condition further participation in this defense program on not being required to engage in such activity that they deem ought be illegal.
It is no surprise that the government is reacting poorly to this. Without commenting on the ethics of AI-enabled surveillance or non-HITL kill chains, which are fraught, I understand why a department of government charged with making war is uninterested in debating this as terms of the contract itself. Perhaps the best place for that is Congress (good luck), but to remind: the adversary that these people are all thinking about here is PRC, who does not give a single shit about anyone's feelings on whether it's ethical or not to allow a drone system to drop ordinance on it's own.
[1] https://media.defense.gov/2026/Jan/12/2003855671/-1/-1/0/ART...
Theoretically, but this would run the risk of collapsing the US tech sector, which at this point is a significant part of the strength of the US economy, and thus making it likely that the Republicans will lose power in the next elections.