[ExI] OpenAI Reaches A.I. Agreement With Defense Dept. After Anthropic Clash

[Keith Henson]

What is inconsistent is that the defense official line is that they
don't want to do mass surveillance or lethal unsupervised lethal
weapons, but they insist on getting an AI, which will do these.

I wonder if Hegseth has considered being a target of such an AI?  I
doubt it; foresight is not a strong trait of the people currently in
power.

Keith

On Sat, Feb 28, 2026 at 1:21 PM John Clark via extropy-chat
<extropy-chat at lists.extropy.org> wrote:
>
> On Sat, Feb 28, 2026 at 10:42 AM <spike at rainier66.com> wrote:
>
>> > Former officials don’t know what is going on in the pentagon any more than the NYT does.  Claude may be good, but you can be sure that whatever they have now is better.
>
>
>
> What?! You believe that the Pentagon has access to a smarter AI than anything the AI companies have?? For that to be true it would mean that for the last several years hundreds of the very top AI researchers in the entire world have decided that they would rather be government employees and work for the military rather than work for Anthropic or OpenAI or Google and become filthy rich. And I don't believe there is a snowballs chance in hell of that has occurred, or will ever occur.
>
> By the way I'm curious if you agree with He Who Must Not Be Named that Anthropic is a "Radical Left AI company" run by "Leftwing nut jobs" because they don't think an AI conducting mass surveillance on the American people or an AI controlling a lethal weapon without a human being in the decision pipeline are good ideas. I suppose you do agree, I suppose you feel that it's your duty to agree with everything he does and says. I'd love it if you told me I'm wrong about that.
>
>
>  John K Clark
>>
>>
> _______________________________________________
> extropy-chat mailing list
> extropy-chat at lists.extropy.org
> http://lists.extropy.org/mailman/listinfo.cgi/extropy-chat