[ExI] AGI is going to kill everyone

Adrian Tymes atymes at gmail.com
Tue Jun 7 23:39:39 UTC 2022

On Tue, Jun 7, 2022 at 4:11 PM BillK via extropy-chat <
extropy-chat at lists.extropy.org> wrote:

> Eli's response is to say that powerful groups like Facebook AI or the
> Chinese don't care about safety. They just want to win the race to get
> super-powerful AGI for themselves. Their assumption is that the first
> AGI will rule the world and they will own it.

Facebook doesn't seem to be in it to conquer the world, though.  Just to
increase their profits, while leaving political power to others.  But
China, maybe.

Eli fears that instead
> their recklessness will destroy the human race.


...is an example of a Pascal's Wager type fallacy; any such future AI could
as easily be one that sees no value in punishment but does reward those who
brought it about, meaning that not thinking about it would be the losing

...is also something that calling out can apparently make those who have
fallen into this particular cognitive trap, flinch.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20220607/77c5b4ca/attachment.htm>

More information about the extropy-chat mailing list