[ExI] How Trump Is Dismantling America’s Scientific Research Machine

[spike at rainier66.com]

-----Original Message-----
From: extropy-chat <extropy-chat-bounces at lists.extropy.org> On Behalf Of Adrian Tymes via extropy-chat
Sent: Saturday, 31 October, 2026 1:02 PM
To: ExI chat list <extropy-chat at lists.extropy.org>
Cc: Adrian Tymes <atymes at gmail.com>
Subject: Re: [ExI] How Trump Is Dismantling America’s Scientific Research Machine

On Sat, Oct 31, 2026 at 3:56 PM spike jones via extropy-chat <extropy-chat at lists.extropy.org> wrote:
> .>> On Behalf Of BillK via extropy-chat
> >…Won't AI  be doing all the scientific research soon?
>
> Ja, and if AI is doing it, why should taxpayers hafta pay for it?

Because someone has to pay for the data centers (construction, and electricity and other costs of operation)...

Ja.  The guy who built it and the stock holders.  Government should have no stake in that at all.  The taxpayer will profit from it however, for it generates revenue which goes to the government.

>...not to mention cleaning up what the AI gets wrong....

It is easy enough to foresee that humans cannot clean up what AI gets wrong: we won't be smart enough.  But AI might be able to clean up what AI gets wrong.

>...Also, the research is of value to the humans a lot more than to the AI....

Sure is.  We humans win in the deal (assuming AI doesn't kill everybody.)

It is easy enough to foresee that medical breakthroughs will be made by AI very soon.

>> The cool part about AI research is that taxpayers aren’t on the hook for it.  Rich guys pay for it all.

>...Demonstrably untrue.  They pay for some of it, but far from all of it...

Solution: don't let government pay for AI research.

>...Also, AI research is only a small fraction of all scientific research...

No problem.  Let AI do AI research and any other area it can take on.  Medicine is a perfect example.  We are held back because we haven't been able to derive a mathematical model for how biological systems work.  That problem is just too complicated for humans.  But until we can derive some kind of mathematical model, even if it is piecewise, applicable to only some subsystems, our medical research is clumsy trial and error, without a unified theory, without universal guiding principles.  We find stuff, but it's by fortunate accident.

More specifically, consider cryonics.  Reading and mapping a frozen brain, creating a working simulation of that brain is a task which I confidently predict is beyond the capacity of unfrozen brains.  For cryonauts to have any chance at all requires AI way beyond what we now have.

By this view, AI is dangerous but simultaneously it is our only hope.

spike  

_______________________________________________