The power of pentagon contracts… Anthropic will bend the knee for the money. If they won't, there are plenty of others that will. Tough spot to be in.

Yep.

Come on BigBear! Tell them you'll do whatever they want if Anthropic won't.

I've seen this play out on TV... It's called "Person of Interest".

Anthropic will go rogue and develop competing AI to battle the evil governmental AI. Though it will morph into something even bigger and will help save individuals the AI has calculated to be in imminent danger!

Pichael Thompson said:

---

My guess is the msm took Hegseth's point way out of context as usual, but I'll wait to see

---

Yep, notoriously hard on Trump Fox News is doing their best to spin this I'm sure...

Less Evil Hank Scorpio said:

---

There is reporting out there now that Anthropic's current position is that they will not comply, but time will tell if they really want to weather that storm.

Their CEO made a very interesting point that an AI system lacks the crucial component of being able to use judgment to refuse an unconstitutional order. I hadn't thought of that before, but that logic could pave the way for this to become a Supreme Court issue at some point.

---

Probably the only time I wish SCOTUS acted proactively. We NEED to get ahead of this coming storm because it is not going to go away. I am of the mind fully AI controlled killchains are inevitable at this point (and probably not even by the U.S.) but the longer we stand against it with ironclad laws, the better.

AI needs to be muzzled from the get go in every avenue, not reacted to after the fact in everything. Which is where we currently are. I hate new laws like everyone else, but no one can argue that this is not different this time around.

If Anthropic doesn't want to do business with DoD, fine.

Labeling them a supply chain risk seems problematic.

I would think AI for defense use would need to be specially developed for that purpose and not a general model.

Lol just have to say I love reading the liberal tears over stuff like this

Yall are extremely naive at the AI capabilities here. They are not asking a chatgpt who to blow up. Lol.

harge57 said:

---

Yall are extremely naive at the AI capabilities here. They are not asking a chatgpt who to blow up. Lol.

---

It's not asking what to blow up. You would use the AI (combined with parameters that you send) to be able to identify things you classify as valid targets. It's easily doable now.

mickeyrig06sq3 said:

---

harge57 said:

---

Yall are extremely naive at the AI capabilities here. They are not asking a chatgpt who to blow up. Lol.

---

It's not asking what to blow up. You would use the AI (combined with parameters that you send) to be able to identify things you classify as valid targets. It's easily doable now.

---

To expand on the "easily doable". It's easy to identify and choose to kill the target. The hard part is making sure your AI knows when not to kill. Let's say we've got a drone outfitted with air-to-ground capabilities. Take AI with access to cell records and facial rec. The IMEI you're looking for lights up on a tower you're monitoring. A drone nearby is able to confirm that the person with the cell phone is the target you've been looking for, launch. Except you didn't program the AI to account that he was going into a mosque, or he was at a hospital, etc.

However, using AI for every step above (except the launch part) is perfectly acceptable. Now I can have 15 drones in the air doing the things above. But at the launch step, it just sends an alert to the airman with the information it's gathered and then the airman brings up the drone and takes it from there.

What an odd notion that a CEO is turning down one of the most lucrative contracts available due to reservations over his own product, and Hegseth is threatening him to proceed anyway.

500,000ags said:

---

What an odd notion that a CEO is turning down one of the most lucrative contracts available due to reservations over his own product, and Hegseth is threatening him to proceed anyway.

---

Probably because if we don't China will roll out a ****tier version sooner.

Can we just let Ukraine test this on Russians for us? And then review performance data of human in the loop decision making vs no human in kill chain?

It's worth considering that the human in the loop might prevent errors (wrongful killings) but might result in preventable deaths because it is presumably slower. The end result is the same in my opinion.

This is from the perspective that "everyone" (china, Russia, and even us one day) will end up doing this. Might as well lead the charge and be the best at it.

That could be a big positive for Anthropic and establish their branding as the safe/ethical AI choice.

Did the DOW not read the terms and conditions of the initial agreement or just decide to eff them terms?

harge57 said:

---

500,000ags said:

---

What an odd notion that a CEO is turning down one of the most lucrative contracts available due to reservations over his own product, and Hegseth is threatening him to proceed anyway.

---

Probably because if we don't China will roll out a ****tier version sooner.

---

This.

Those who believe that AI leading to our complete destruction is coin-flip odds right now don't believe that because they think it's just become so smart it will overtake us and turn us into batteries ala The Matrix.

They believe a Cold War scenario is inevitable where multiple countries are racing to upgrade their systems and "can't fall behind!" which will lead them to have less and less guardrails, which will lead to automated decisions that lead us down the road to destruction.

And right now the only thing that makes that less likely is civil unrest and countries cannibalizing themselves when AI and robotics/automation does away with more than half the jobs.

So it's well on it's way to killing us all - we're just not sure of the timetable yet

Anthropic's AI is very left-wing coded. I'm guessing this came up because the Pentagon asked it to do something completely legal, and it refused.

Less Evil Hank Scorpio said:

---

At high stakes Pentagon meeting today Sec Hegseth gave Anthropic head Dario Amodei ultimatum to allow the Pentagon to use Anthropic’s AI model for mass domestic surveillance and kinetic autonomous operations without human oversight or face censure and be labeled “supply chain…

— Jennifer Griffin (@JenGriffinFNC) February 24, 2026

FNC is reporting there was a meeting today in which Hegseth demanded the two restrictions imposed by Anthropic on their models used by the gov't be lifted. Those two restrictions are 1) no mass surveillance and 2) no "autonomous kinetic operations", in other words a human has to be involved and not just let the model decide who to kill.

Here is the quote from the tweet:

Quote:

---

At issue is Anthropic's two stipulations that its advanced AI model currently used in the Pentagon's classified systems is NOT used for autonomous kinetic operations (Anthropic currently requires human oversight of autonomous operations when used to kill things for safety reasons because they don't know how the autonomous system will react and could even endanger soldiers using the model; soldiers and others could lose control of the model and automatically start killing large groups without humans in the "kill chain.") Second Anthropic bars its models from being used for mass domestic surveillance. Hegseth wants these restrictions lifted.

---

Demanding those two restrictions be lifted is extremely troubling. Threatening Anthropic if they don't lift those restrictions is also troubling.

---

Did you ever consider that the enemies of the USA are developing advanced AI without these restrictions? IMO, we absolutely must be able to do mass surveillance and autonomous kinetic operations. I don't like it. I think things could go wrong. But, THIS IS COMING. We MUST be ready and willing to do this because our adversaries won't blink at the chance to do it.

ErnestEndeavor said:

---

Anthropic's CEO is one of the biggest proponents of AI safety initiatives. He's seen what goes on in tests behind the scenes and it freaks him out.

What he doesn't want to see is some autonomous system going rogue or hallucinating and taking out the wrong group of people.

---

I dunno about you but I'm skeptical of anyone who says they are a big proponent of AI safety that is also working for the DoD. That dog don't hunt.

Houston Lee said:

---

Less Evil Hank Scorpio said:

---

At high stakes Pentagon meeting today Sec Hegseth gave Anthropic head Dario Amodei ultimatum to allow the Pentagon to use Anthropic’s AI model for mass domestic surveillance and kinetic autonomous operations without human oversight or face censure and be labeled “supply chain…

— Jennifer Griffin (@JenGriffinFNC) February 24, 2026

FNC is reporting there was a meeting today in which Hegseth demanded the two restrictions imposed by Anthropic on their models used by the gov't be lifted. Those two restrictions are 1) no mass surveillance and 2) no "autonomous kinetic operations", in other words a human has to be involved and not just let the model decide who to kill.

Here is the quote from the tweet:

Quote:

---

At issue is Anthropic's two stipulations that its advanced AI model currently used in the Pentagon's classified systems is NOT used for autonomous kinetic operations (Anthropic currently requires human oversight of autonomous operations when used to kill things for safety reasons because they don't know how the autonomous system will react and could even endanger soldiers using the model; soldiers and others could lose control of the model and automatically start killing large groups without humans in the "kill chain.") Second Anthropic bars its models from being used for mass domestic surveillance. Hegseth wants these restrictions lifted.

---

Demanding those two restrictions be lifted is extremely troubling. Threatening Anthropic if they don't lift those restrictions is also troubling.

---

Did you ever consider that the enemies of the USA are developing advanced AI without these restrictions? IMO, we absolutely must be able to do mass surveillance and autonomous kinetic operations. I don't like it. I think things could go wrong. But, THIS IS COMING. We MUST be ready and willing to do this because our adversaries won't blink at the chance to do it.

---

"Mr. President, we must not allow a mine shaft gap!"

Drahknor03 said:

---

Anthropic's AI is very left-wing coded. I'm guessing this came up because the Pentagon asked it to do something completely legal, and it refused.

---

The debate is about Anthropic usage policy…

Anthropic has concerns about its software being used for mass surveillance or autonomous targeting.

The concern is largely based on issues of being held liable if the software isn't good enough.

DoD on the other hand says : "if you sell it to us we can do what we want with it and if you don't, we're going to label you a supply chain risk."