Our agreement with The Department of War

OpenAI: "We think our agreement has more guardrails than any previous agreement for classified AI deployments, including Anthropic’s."

Then why did the DoD kick out Anthropic? Ah, I see...

OpenAI Red Line #1: "No autonomous weapons." The contract language: AI won't direct weapons "in any case where law or Department policy requires human control." Wherever policy doesn't require human oversight, the restriction disappears.

OpenAI Red Line #2: "No mass domestic surveillance." The contract language: no "unconstrained monitoring... consistent with these authorities" - authorities that have historically been used to justify enormous surveillance programs (see: FISA, EO 12333). They're not prohibiting surveillance.

OpenAI Red Line #3: "No high-stakes automated decisions." The contract language: AI won't make decisions "that require approval by a human decisionmaker under the same authorities." If the government changes what requires human approval, the guardrail moves with it.

Bottom line is OpenAI built a framework that sounds like constraint but preserves maximum flexibility for the government. Every red line contains an escape clause written by the entity being restrained.

There is no real guardrail here. Sam Altman is playing word games like a politician.

 

Do we have actual guardrails on autonomous cheetah robots, or are we officially doomed?

 

OpenAI is why the Middle East is burning.

“you’re right that was a school. We will do better next time”

 

This breh crashed out on Nazi Twitter last night after randomly digging the hole deeper in some AMa style clusterf



You hate to see it

 

"go do something for a while."
My students sometimes write that way, but not by the time they are seniors.

 

I have some presentations etc for our C-Suite AI huffers this week and I plan on answering their questions with this if I start mucking up

 

C-Suite AI huffers

Going to steal that one.

 

TACO (trump chickening out again), as it relates ti Anthropic's Claude AI, being played out in the Iran War.

on Jun 3 2026, Trump 2.0 blacklists Anthropic as AI firm refuses Pentagon demand

• Trump ordered U.S. government agencies to “immediately cease” using technology from the artificial intelligence company Anthropic.
• DUI Hegseth, soon after Trump’s order, said he was ordering the Pentagon to “designate Anthropic a Supply-Chain Risk to National Security”

against this backdrop, CBS is reporting that

the Defense Department uses Claude for synthesizing documents and making logistics and supply chains more efficient, among other tasks

 

In the case of Anthropic v. DOD, the DOD's lawyer is saying this X post from Hegseth is just a social media and DOD has no intent to do what he stated.

In conjunction with the President’s directive for the Federal Government to cease all use of Anthropic’s technology, I am directing the Department of War to designate Anthropic a Supply-Chain Risk to National Security. Effective immediately, no contractor, supplier, or partner that does business with the United States military may conduct any commercial activity with Anthropic.​

Funny - Fox News' lawyers once successfully argued Tucker Carlson was just an entertainer and no reasonable person should take him literally. Now the DOD is making the same argument about their own Secretary. Tucker was a cable host. Hegseth is running a department currently fighting a war, and apparently, we should not take his words seriously.

​

 

A federal judge has ruled against the administration in its effort to punish Anthropic for speaking out. The excerpt below is worth reading. Not just for what it reveals about this case, but for what it confirms about this administration's relationship with the First Amendment.

Not an isolated incident. It's a pattern.

What the court calls "classic illegal First Amendment retaliation" is, by now, a classic administration tactic.


Judge blocks Pentagon from labeling Anthropic AI a "supply chain risk" and halts Trump's ban on federal use - CBS News

The record supports an inference that Anthropic is being punished for criticizing the government’s contracting position in the press. In their announcements, the President and Secretary Hegseth called Anthropic “out of control” and “arrogant,” describing its “sanctimonious rhetoric” as an attempt to “strong-arm” the government. The Department of War’s records show that it designated Anthropic as a supply chain risk because of its “hostile manner through the press.” Punishing Anthropic for bringing public scrutiny to the government’s contracting position is classic illegal First Amendment retaliation.

Moreover, Defendants’ designation of Anthropic as a “supply chain risk” is likely both contrary to law and arbitrary and capricious. The Department of War provides no legitimate basis to infer from Anthropic’s forthright insistence on usage restrictions that it might become a saboteur. At oral argument, government counsel suggested that Anthropic showed its subversive tendencies by “questioning” the use of its technology, “raising concerns” about it, and criticizing the government’s position in the press. Nothing in the governing statute supports the Orwellian notion that an American company may be branded a potential adversary and saboteur of the U.S. for expressing disagreement with the government.

 

TACO (trump chickening out again), as it relates to Anthropic's Claude AI, being played out in the Iran War.

against this backdrop, in March 2026, CBS has been reporting that
the Defense Dept uses Anthropics' Claude for synthesizing documents and making logistics and supply chains more efficient


the latest reporting, as of 4-20-26, is that

Despite a Pentagon blacklist, the NSA is using Anthropic's Mythos

According to unconfirmed reports, Pentagon officials added Anthropic to their restricted vendor list after discovering Beijing-linked funding in the AI company’s investment portfolio. The blacklist theoretically prevents any Defense Department entity from purchasing or deploying Anthropic’s products. Yet NSA operations teams allegedly maintain access through:

• Third-party contractors
• Pre-existing licenses that predate the restriction

​

This mirrors the classic Trump 2.0 move of banning TikTok while half the staff still scrolls during lunch breaks.​