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

John Clark johnkclark at gmail.com
Mon Mar 2 13:06:37 UTC 2026


On Sun, Mar 1, 2026 at 6:03 PM <spike at rainier66.com> wrote:

 *>>> **John you persist in getting caught up in politics. *
>>
>>

>>… *And  STILL no answer to my very simple question. You just can't say it*
>> *….*
>
>
>
> *> OK.  What was the question? *
>

*Spike, do you really expect me to believe you don't know the question
after I've repeated it twice? You realize that what POTUS said was nonsense
but you are unable to say so, I suppose you feel that any criticism of the
chief of your tribe would be a betrayal. My tribe is science so I am free
to criticize him when he does something stupid, which is often, and praise
him when he does something right, which is rare. It's so rare it comes
close to thermodynamic impossibility but not quite, it is observable.  *

>
*> Good for him.  But the military still can’t use AI with guardrails. *
>
*A man who has a long history of telling thousands of lies a year says he
would never do something so guardrails are unnecessary, but for some
unstated reason the man absolutely insists that he must possess the ability
to do that thing that he would never do anyway. Are you really so naïve
that you believe him? I don't think so, I could be wrong but I think a more
probable theory is you feel that denying the man anything he wants would be
betraying your tribe. *


> * > They need to know for sure their AI weapons will fire when ordered to
> do so.*
>

*Then they should use their own wonderful AI which you claim is much much
better than Claude or anything else that civilians have because it's the
military that is at the cutting edge of AI research not the AI companies.
But... If the military really did have an AI that was that superior then I
don't understand why the Singularity hasn't already happened. And I don't
understand how humans can EVER "know for sure" that their AI will do what
they order it to do.  *



> *>>A trivial little thing like violating the Constitution is not going to
>> stop You Know *
>
>
>
> *> I see, so you presume guilt, then from that derive guilt?*
>

*I don't have a clue what you mean by that. *

>

> *> The mass surveillance by government is illegal under the constitution. *
>

*If you had actually watched that very important interview with Dario
Amodei where he explains why he did what he did, you would know that he was
NOT talking about what was and was not constitutional, he was talking about
what was and was not Orwellian. I don't think Amodei even mentioned the
Constitution, which isn't surprising because it isn't relevant; none of the
18th century framers of the Constitution knew the first thing about AI, and
our current POTUS doesn't know much more.  *

* John K Clark*

>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20260302/ae1e9bc6/attachment.htm>


More information about the extropy-chat mailing list