Our Agreement with the Department of War

> For intelligence activities, any handling of private information will comply with the Fourth Amendment, the National Security Act of 1947 and the Foreign Intelligence and Surveillance Act of 1978, Executive Order 12333, and applicable DoD directives requiring a defined foreign intelligence purpose. The AI System shall not be used for unconstrained monitoring of U.S. persons’ private information as consistent with these authorities. The system shall also not be used for domestic law-enforcement activities except as permitted by the Posse Comitatus Act and other applicable law.

My reading of this is that OpenAI's contract with the Pentagon only prohibits mass surveillance of US citizens to the extent that that surveillance is already prohibited by law. For example, I believe this implies that the DoW can procure data on US citizens en masse from private companies - including, e.g., granular location and financial transaction data - and apply OpenAI's tools to that data to surveil and otherwise target US citizens at scale. As I understand it, this was not the case with Anthropic's contract.

If I'm right, this is abhorrent. However, I've already jumped to a lot of incorrect conclusions in the last few days, so I'm doing my best to withhold judgment for now, and holding out hope for a plausible competing explanation.

(Disclosure, I'm a former OpenAI employee and current shareholder.)

> The Department of War may use the AI System for all lawful purposes, consistent with applicable law, operational requirements, and well-established safety and oversight protocols. The AI System will not be used to independently direct autonomous weapons in any case where law, regulation, or Department policy requires human control, nor will it be used to assume other high-stakes decisions that require approval by a human decisionmaker under the same authorities. Per DoD Directive 3000.09 (dtd 25 January 2023), any use of AI in autonomous and semi-autonomous systems must undergo rigorous verification, validation, and testing to ensure they perform as intended in realistic environments before deployment.

The emphasized language is the delta between what OpenAI agreed and what Anthropic wanted.

OpenAI acceded to demands that the US Government can do whatever it wants that is legal. Anthropic wanted to impose its own morals into the use of its products.

I personally can agree with both, and I do believe that the Administration's behavior towards Anthropic was abhorrant, bad-faith and ultimately damaging to US interests.

Not great? Seems kind of loose language? It isn't OpenAI saying no autonomous weapons use, but only that use must be consistent with laws, regulations, and department policies: "The Department of War may use the AI System for all lawful purposes, consistent with applicable law, operational requirements, and well-established safety and oversight protocols. The AI System will not be used to independently direct autonomous weapons in any case where law, regulation, or Department policy requires human control, nor will it be used to assume other high-stakes decisions that require approval by a human decisionmaker under the same authorities."

More of the same here. Not a wonder why the DoD signed with OpenAI and instead of Anthropic. Delegating morality to the law when you know the law is not adequate seems like "not a good thing".

"For intelligence activities, any handling of private information will comply with the Fourth Amendment, the National Security Act of 1947 and the Foreign Intelligence and Surveillance Act of 1978, Executive Order 12333, and applicable DoD directives requiring a defined foreign intelligence purpose. The AI System shall not be used for unconstrained monitoring of U.S. persons’ private information as consistent with these authorities. The system shall also not be used for domestic law-enforcement activities except as permitted by the Posse Comitatus Act and other applicable law."

Saying that an entity with the power to make its own laws can use something for "all lawful purposes" is saying they can use it for anything.

How incredibly unsurprising. This is why it is pointless to make moral stands as employees when you do not ultimately have power over the companies decisions. The only power you have is to quit.

I wonder how many will do so, and how many will simply accept Sam’s AI written rationalization as this own and keep collecting their obscene pay packages…

I don't think Anthropic is a saint that will never do anything unethical. I don't think ChatGPT is any better or worse.

But I do think my cancelling ChatGPT so I can try Claude, at this time, sends the message I want to send, which is why I did it.

OpenAI: "let's delegate morality to laws that we know are wholly inadequate for AI to absolve ourselves of any moral responsiblity."

> Fully autonomous weapons. The cloud deployment surface covered in our contract would not permit powering fully autonomous weapons, as this would require edge deployment.

Can anyone explain this constraint?

Why do fully autonomous weapons require edge deployment?

Does "fully autonomous" in this context mean "disconnected from the Internet"?

If so, can a drone with Internet connectivity use OpenAI?

Or maybe it's about on-premise requirements: the military doesn't want to depend on OpenAI's DCs for weaponry, and instead wants OpenAI in their own DCs for that?

As a stealth ceo of a profitable SaaS. This is a nice reminder for my company to wind down its relationship with OpenAI. I have no doubt Anthropic will eventually become evil but at least they have a backbone today.

Goodbye Sam.

Edit: Also, referring to the DOD as the Department of War is cringe.

It's hard to believe that this was written in any good faith when there's so much beating around the bush and careful legalese wordplay.

I look forward to seeing more abusive tactics by the US government powered by AI and the language OpenAI will use to confuse the public into thinking they aren't responsible.

“The Department of War may use the AI System for all lawful purposes, consistent with applicable law, operational requirements, and well-established safety and oversight protocols.”

So DoW did get the “all lawful purposes” language they were after, with reference to existing (inadequate, in my view) regulations around autonomous weapons and mass surveillance.

Does OpenAI enforce those red lines in all contracts?

From what I can tell the Anthropic issue was triggered by something Palantir was doing as a contractor for DoW, not anything related to direct contracts between DoW and Anthropic, and DoW was annoyed that Anthropic interfered with what Palantir was up to.

In other words will OpenAI enforce these "red lines" against use by a third-party government contractor?

If not, this seems pretty meaningless if they are essentially playing PR while hiding behind Palantir.

These communications offend me because they treat the audience like they’re stupid, stupid, stupid.

But I imagine that being honest about your corporate identity is suboptimal. It’s probably an important cognitive dissonance tool for the employees? It’s like when autocracies repeat big obvious lies endlessly. Gives those who want to opt out of reality an option.

In my opinion all this discussion of the contract language is a subterfuge. The real question is why the government was requesting this language in the first place. Clearly there’s more to it than a legal battle.

In my mind, the government would be fully happy to use this to surveil citizens (and indeed anyone) with or without any legal basis, but the issue was that Anthropic has a safety stack / training and inference protocols that it follows. Refusals, abuse models, and manual guardrails. They didn’t want to shut those off. Likely there were some very basic technical reasons, some being that the team’s safety posture is fully ingrained in the model itself and thus difficult to remove.

In this document, OpenAI admits that while they are not “turning off” their safety stack, they are completely willing to provide the government with a different model, different guardrails, etc. That should be incredibly concerning. Anthropic was unwilling to do this, cited their ToS, and ultimately had to walk away from the deal. Given that the government (DoW really) framed this in terms of a hilariously stupid position (surveillance and autonomous weapons), Anthropic felt that this was something they could voice to the public and therefore the entire guardrails discussion turned into a “we want the language changed”. Also the government can’t actually compel Anthropic to create new guardrails so they had no choice but to raise the stakes, make this a moral thing, and basically accuse Anthropic of being woke.

IMO this is really sad for OpenAI employees. Yet again Sam Altman proves that he wants to weasel his way around public perception. Folks at the company have to grapple with working for someone of that disposition.

Not saying it was, but the course of actions awfully look like a setup was made for Anthropic.

This blog post really doesn't make it sound any better there is no clear refusal to participate in the questionable uses Anthropic was against. Merely must be legal and must be tested.

This feels like IBM in the 1930s selling tabulating machines to the Germans and downplaying their knowledge of their use. They seem to want us to naively believe they won't use it for exactly what the military has always wanted, autonomous weapons and mass surveillance. Further more there are much more mundane use they might make of the technology that is perfectly legal yet morally in gray areas.

"What if the government just changes the law or existing DoW policies?"

Our contract explicitly references the surveillance and autonomous weapons laws and policies as they exist today, so that even if those laws or policies change in the future, use of our systems must still remain aligned with the current standards reflected in the agreement.

So, this apply only if they changes the law, not if they break the law.

"What happens if the government violates the terms of the contract?"

As with any contract, we could terminate it if the counterparty violates the terms. We don’t expect that to happen.

WE COULD [...]. Yeah, I believe

I deleted my OpenAI account months ago. If LLMs and adjacent technology are truly a paradigm shift, I can’t think of many worse than Sam Altman to shepard us through that. He is a pure opportunist who has already shown how little he believes in outside of his own power and wealth.

This is the same company that started as a nonprofit dedicated to open AI safety research, then became a capped-profit entity, then effectively closed-source, then dropped the cap, and is now pursuing full for-profit conversion. Every single guardrail they've set for themselves has been quietly revised or removed once it became inconvenient. Anyone want to bet on how long those exclusions last?

Remember that this is the future that Altman is building for “all of humanity”

The agreement puts no restrictions on the government beyond “all lawful purposes,” which is what Anthropic objected to.

> “ The Department of War may use the AI System for all lawful purposes… [proceeds to describe current law, with clear openings if the law changes]”

Thus, OAI is relying on the Trump administration’s interpretation of current law. Which, I will remind readers, suggests that it is legal to kill civilians on boats, kidnap foreign leaders, deploy troops in American cities, shoot American citizens protesting ICE.

Yeah I’ve cancelled my OAI sub.

Ah, yes, OpenAI, org known for keeping the word they gave on the direction of the company, with literal lie about that in their very name.

> Why could you reach a deal when Anthropic could not? Did you sign the deal they wouldn’t? Based on what we know, we believe our contract provides better guarantees and more responsible safeguards than earlier agreements, including Anthropic’s original contract.

Weak. You reached a deal that Anthropic could not because you demanded more safeguards than Anthropic?? (Based on what you know, of course).

Makes total sense!

Added to the ever growing commercial product shit list.

I’m going to be left with scrap PCs and Debian at this rate.

OAI: “If they stretch, reinterpret or beak the law with our systems, well, that’s on them. Good luck everybody!”

I wonder if the autonomous weapon platforms they'll build will be surprisingly susceptible to friendly fire... I don't think the DoW knows what kind of Pandora's Box they just bought.

Related: https://news.ycombinator.com/item?id=47197505

Are they not allowed to say department of defence? I know botj names are official now but this is a choice on their own blog.

I don't really have anything against OpenAI's stance here. If that's how they want it to be, they have that choice.

But Sam pretending that he wanted the same restrictions as Anthropic *and* seeing how quickly they swooped in and made a deal with the DoD really skeeves me out. (But Sam always gave me the heebie jeebies).

Anyway, I've always preferred Claude, so I'm going to happily stay a paying customer there. This may end up being a big "branding" differentiator.

The timing of the release and the phrasing used in the headline: Woof.

Wow, incredibly anti human from Sam. Humanity's only hope seems to be Anthropic getting to ASI first and locking OpenAI out.

You’re done, Sam.

Sam won't even sign his name to this press release.

time to delete my account.

You’re done Sam.

> AI-enabled mass surveillance is fine as long as it isn’t domestic.

> We want AI to be aligned with all of humanity.

One of many contradictions. Liars.

Why is everyone mad if they have better guaranties that anthropic use to have ?

I feel like I keep saying this but it's critical to remember what OpenAI says on its blog doesn't have to align with what it delivers to the Pentagon.

If I hadn’t already canceled my account over them including ads in a paid service, I’d certainly be canceling over this. Anthropic is lucky they have some spine, otherwise they’d have been binned as well.

All this says is that all uses must remain lawful. So what? As if this admin has been a shining example of lawful behavior.

This is weak.

Well worded. Plentiful protections for themselves and others.

by now, we all know the core characters of altman and trump and their enablers. press releases (hell any of their words) mean nothing. they are just distracting fodder for fools and sycophants.

Here is a point Mr. Altman might not have considered. Everyone in Trump's circle will probably get a pardon no matter what. but not the CEOs who were collaborators. not in the inner circle but still complicit.

Even Google and Microsoft should be worried. This is like 1936 germany, we have ways to go. Look at the tune this administration is singing, if they get their way these CEOs aren't looking at law suits and federal investigations, the current order of things will be long gone by the time people start asking who's responsible for all the blood on the streets.

It’s the fucking department of defense.

More Sam Altman lies. Can’t believe anything that jerk says

OpenAI basically bribed the government into attacking Anthropic, via political donations to the MAGA PAC. They couldn’t not compete with an inferior product so Altman and Brockman went this route.

As for OpenAI’s defense - not buying it.

“OpenAI’s President Gave Millions to Trump. He Says It’s for Humanity”: https://www.wired.com/story/openai-president-greg-brockman-p...

> > Do you think Anthropic should be designated as a “supply chain risk”?

> No, and we have made our position on this clear to the government.

Look, this is the most important thing that everyone needs to understand: Your opinion on this is not welcome here. Your opinion on how the government uses the tools it purchases are unimportant and a non-factor. It is not appropriate for you to share your opinion on this. The government that was elected by the people is the sole decision maker. That’s the agreed social norm that we have in this country. What you’re doing is a minor subversion of our democratic republic, even if it feels like you’re standing on firm moral ground.

The DoD can and will deploy eye watering amounts of capital in the pursuit of its mission. That mission includes artificial intelligence based war systems. If you want a piece of that pie, even indirectly, you need to shut the fuck up and kiss the ring. That’s the reality. You don’t have to like this, but you’re shockingly naive if you didn’t know the world worked this way. The DoD spends nearly a trillion dollars a year, did you really think that was entirely spent on raw materials?

Their systems will be built to their spec, one way or another. They will seize your source code and training sets. They will build data centers. Nothing can stop this. People are making this about Trump and Hegseth, but it’s bigger than that. This transcends political parties. Obama’s DoD would make the same stand, and you’re naive if you don’t think so. Our war machine never loses in the game of politics.

Do we really need to read the text of a statement entitled "Our agreement with the department of war"? If it weren't the US, it would still be something that any person of moral character would never get in the position to write.

And it _is_ the US department of war - just now entered into yet another war of aggression against Iran, with no cause nor legal basis (not even domestic IIANM), in and endless list of wars, direct and indirect. With another crown jewel being the support, funding and arming for the still-unhalted genocide in Gaza.

now DeepSeek and Qwen obtain similar or even more lenient terms, then a reckless slippery slope for supremacy and maybe at some point there won't be 2 player fighting, but a 3rd created by this exact dynamic, an autonomous unaligned undetected AI

too late bro