Path: utzoo!attcan!uunet!mcvax!ukc!dcl-cs!simon
From: simon@comp.lancs.ac.uk (Simon Brooke)
Newsgroups: comp.ai
Subject: AI seen as an experiment to determine the existence of reality
Summary: If our current understanding of the world is right, AI should work
Keywords: Reality, brain, computability
Message-ID: <517@dcl-csvax.comp.lancs.ac.uk>
Date: 9 Jun 88 09:06:42 GMT
References: <1313@crete.cs.glasgow.ac.uk> <1053@cresswell.quintus.UUCP> <1335@crete.cs.glasgow.ac.uk>
Reply-To: simon@comp.lancs.ac.uk (Simon Brooke)
Organization: Department of Computing at Lancaster University, UK.
Lines: 29

Following the recent debate in this newsgroup about the value of AI, a
thought struck me. It's a bit tenuous....

As I understand it, Turing's work shows that the behaviour of any
computing device can be reproduced by any other. Modern cosmology holds
that:
	1] there is a material world.
	2] if there is a spiritual world, it's irrelevent, as the
		spiritual cannot affect the material.
	3] the brain is a material object, and is the organ which largely
		determines the behaviour of human beings.

If all this is so, then it is possible to exactly reproduce the workings
of a human brain in a machine (I think Turing actually claimed this, but I
can't remember where).

So AI could be seen as an experiment to determine whether a material world
actually exists. While the generation of a completely successful
computational model of a human brain would not prove the existence of th
material, the continued failure to do so over a long period would surely
prove its non-existence... wouldn't it?


** Simon Brooke *********************************************************
*  e-mail : simon@uk.ac.lancs.comp                                      * 
*  surface: Dept of Computing, University of Lancaster,  LA 1 4 YW, UK. *
*                                                                       *
*  Neural Nets: "It doesn't matter if you don't know how your program   *
***************  works, so long as it's parallel" - R. O'Keefe **********