[ExI] Review of AI 2040 Plan

[spike at rainier66.com]

> On Behalf Of BillK via extropy-chat
,,,
>.... If industry develops AI, the two primary goals are 1) making money and
2) making more money.
> I'll taaaaake... Industry for 500 please Alex.   spike
> _______________________________________________


>...AI development is not yet a problem.
It is the no-holds-barred race for AGI that threatens to end humanity.
Improvements to AI can help humans.
But once AGI appears, humans will no longer be in charge.
The AI 2040 plan suggests that a worldwide pause in AI development might
give humans time (aided by AI) to find a path to AGI that doesn't lead to
the end of humanity.

>...But Gemini thinks such a worldwide pause is very unlikely to be
achieved.
BillK
_______________________________________________



I agree with Gemini.  I see no practical means of pausing worldwide
development of SAI.  I also agree we don't know how it will be aligned.  We
can only hope it likes us, or finds some use for us.  I predict we will get
along with it, but have no way to guarantee that.

BillK, a couple of years ago we had a national level election in which one
of the candidates had been assigned the AI Czar.  The AIC made it clear that
her main interest in AI was that it aligns with her party, against the other
major party.

Here's the British press take on it:

https://www.dailymail.com/news/article-12294035/Kamala-Harris-says-artificia
l-intelligence-kind-fancy-thing.html

The Brits have it right: have three major parties rather than two.  Then the
way to win is to achieve consensus of two of those parties.

spike