From: Slawomir Paliwoda (velvethum@hotmail.com)
Date: Thu Oct 28 2004 - 17:14:38 MDT
Marc Geddes wrote:
> You can understand the dilemna of someone who agrees
> with basic Singulatarian concepts, but has serious
> doubts about the specific approach of Sing Inst...
I don't believe that an average Friendliness enthusiast can claim to have
serious doubts about specific approach to FAI if he or she doesn't fully
comprehend what that approach entails exactly, how it all works and why.
Perhaps there are people on this list qualified enough to make a legitimate
assessment of the quality of the theory, but not many. Unfortunately, I
haven't seen these people make any endorsements of SIAI's approach. If
you're out there, please come forward and share your views.
As for vast majority of sl4-ers who lack the qualifications to make that
judgment, whether the approach is safe or not should still be a coin toss. I
realize how easy it is to lose oneself in the uplifting vision of Friendly
post-Singularity, but, without real knowledge necessary to evaluate
potential consequences of SIAI's "approach," it makes more sense to remain
agnostic about that approach's real value.
There's no question that SIAI's work on FAI *theory* is important and should
go on, but unless Seed AI team translates Friendliness into a mathematical
model and proves that it would be impossible for the entity to ever turn
unFriendly, we should be extremely nervous about any plans to *implement*
that theory. The stakes are simply too high.
Slawomir
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:49 MDT