Re: 6 points about Coherent Extrapolated Volition

From: Heartland (velvethum@hotmail.com)
Date: Sun Jul 24 2005 - 14:07:49 MDT


p3 wrote:
> And so, the one weak, fragile, human question I dare
> ask in the wake of all this confusing erudite
> prattling: If CEV itself does not mess around with my
> life, will its *results* mess around with my life? In
> a 1000 years from now, will I be able to choose for
> myself what I want to do with my life? I may not be
> asking these questions if I had actually read the CV
> document, and I apologize for my ignorance. But I am
> worried here, because, quite frankly, "Collective
> Volition" sounds like a monster movie.

My understanding of CV is that it's a decision making mechanism that will
guide actions of the first self-improving capable AI. That first AI with CV
at its core is meant to do one thing only, namely, to produce a Friendly AI.
Whatever the entity CV-driven AI creates, *that* entity will actually be
altering reality in, hopefully, optimally Friendly ways. Once this maximally
Friendly entity exists, CV-driven AI will become obsolete. HTH.

Slawek



This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:51 MDT