Re: How hard a Singularity?

From: James Higgins (jameshiggins@earthlink.net)
Date: Sat Jun 22 2002 - 23:42:58 MDT


At 11:16 PM 6/22/2002 -0500, you wrote:
>James, I have explained why an AI can run on a fairly modest hardware
>platform. I have explained why this AI would be much more intelligent
>than a human. I have explained why the Singularity is VERY near. Eliezer
>has explained why a hard take off is VERY likely. You argue against all
>of these points without any logic.
>
>James, who are you working for?

Sorry, I thought I was providing my logic in my posts. I suppose I'll have
to endeavor to do better in the future.

1) Why an AI can run on a fairly modest hardware platform

Well, I missed that one. If you can point me at it I'll give it a
read. I'm always open minded, but the opinion of many people (some of whom
I believe are very knowledgeable in this area) seems to oppose this point.

2) Why this AI would be much more intelligent than a human

And I have explained why it may not be any more intelligent than a
human. Besided "human-equivalent" would seem (at least to me) to indicate
a level equivalent to a human. Not a super-human. So if all your
statements were true, then the AI would be at an earlier point, thus making
it human-equivalent.

3) Why the Singularity is VERY near

Sorry, I just don't buy that one. Well, unless your definition of VERY
near is at least 5+ years, most likely 10+.

I guess I just don't "get it".

James Higgins



This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:39 MDT