Steve HermanAnthropic draws a line in the sand — stating it won’t allow DoD to use its AI technology for mass domestic surveillance or autonomous lethal weapons. https://www.anthropic.com/news/statement-department-of-war
A US defense undersecretary, Emil Michael, verbally attacks Anthropic CEO Dario Amodei as a liar who has a God-complex. “He wants nothing more than to try to personally control the US Military and is ok putting our nation’s safety at risk.”
James M.@newsguyusa it's remarkable just how *precisely* he's projecting, yet again.
Mike Fraser@newsguyusa I'm pretty sure I heard they were already backing away from their "stand" and giving in to the DODs demands.
FeloniousPunk@mike @newsguyusa That was before the announcement today.
notsoloud@newsguyusa
"The Department of War will ALWAYS adhere to the law"
Which says it's the Department of Defense.
Passwordsarehard4@newsguyusa if you don’t want to “bend to a for profit company” maybe you should do your damn job yourself instead of subbing it out.
Randy Ridenour@newsguyusa @igb The DoD is acting as if Anthropic is the only AI company in the US. They’re perfectly free to contract with someone else.
@newsguyusa People in speedboats want to have a word.
@newsguyusa They’re going to get caught in one year helping DoD with this.
@ClickyMcTicker @newsguyusa i don't think i will give anthropic benefit of the doubt - anthropic wants their hands to be publicly clean, I have no illusions that they would give the US DOD exactly what they want behind closed doors, if enough $ and legal waivers are offered.
@crouton @ClickyMcTicker @newsguyusa I’m not quite so cynical. Sure it’s PR: it’s a press release. But they did say no and damn the consequences.
My guess: they walked into the meeting with Hesgeth hoping to extract gov’t money to make their killing machines infallible. An impossible task, but lucrative in the meantime. Hesgeth is nuts. He wants to kill _now_. They decided that wouldn’t do them any good at all.
Maybe not moral, but sensible.
DB@newsguyusa Most people are not aware of this, but there is a compulsory license statute for copyrighted or patented products (enacted during WWI). So Hegseth could just take their product and do whatever tf he wants with it. He’ll need services as well to do his dastardly deeds, thus the pressure campaign
Nicole Parsons"Anthropic expects large government contracts to be offered via a third party proxy."
A bribe of some kind is being negotiated.
Oblomov@Npars01 @newsguyusa yep, that's exactly what I was looking for. Needs to keep the face clean while the hands get dirty
Bogdan Buduroiu> But today, frontier AI systems are simply not reliable enough to power fully autonomous weapons.
So not today, but nothing stops Anthropic from making fully autonomous weapons once they've gathered enough "training data"
> But using these systems for mass domestic surveillance is incompatible with democratic values.
Neither denounces partially autonomous mass surveillance nor closes the door on AI-driven foreign mass surveillance
This isn't the win people think it is, it's more American exceptionalism at the expense of the Global South.
Fuck Anthropic
@budududuroiu @newsguyusa yeah totally fuck Anthropic.
But in a “normal” world discussions between the gov’t and their military contractors do not make the news. I’d love to see a blowup like this in the UK, where Palantir have got our (Labour!) government hooked on the Koolaid.
BTW - wasn’t this public before the Anthropic PR? Where did the initial report come from?
@Kynx @newsguyusa that's true, from what I heard online (and also personally believe), Anthropic is very ideologically motivated.
I think with the UK it's quite different because of how captured the entire state is by the US. I made this point a couple days ago about how Whitehall is effectively captured by Wall Street.
@budududuroiu @newsguyusa yeah that’s where I first read it: https://www.politico.com/news/2026/02/26/incoherent-hegseths-anthropic-ultimatum-confounds-ai-policymakers-00800135
It’s been rumbling away for a bit.