Some new ideas on Friendly AI

From: Ben Goertzel (ben@goertzel.org)
Date: Mon Feb 21 2005 - 20:44:42 MST


Hi all,

Spurred on by recent discussions & some conversations w/ my wife, I've had
some new thoughts on our familiar topics (Friendly AI, etc.).

See this document,

"Encouraging a Positive Transcension via Incrementally Self-Improving,
Theorem-Proving-Based AI"

http://www.realai.net/ITSSIM.pdf

if you're curious.

I don't claim to have solved the uber-problem of Friendly AI (yet), but the
ideas in this paper indicate the sort of direction that I think is most
likely to be useful in terms of moving toward a solution.

This, like each of my previous essays on Friendly AI related topics (most
linked to via www.goertzel.org/dynapsyc/ , only tells part of the story as I
see it. After I finish the five other books in my queue I'll probably
weave all these essays together with some other futurist material I have
lying around, in to a book on "Encouraging a Positive Transcension." But
that may take a while unless my schedule frees up a lot, as I have a lot of
things to do besides write books these days ;)

-- Ben



This archive was generated by hypermail 2.1.5 : Tue Feb 21 2006 - 04:22:53 MST