Newsgroups: comp.ai.philosophy
Path: cantaloupe.srv.cs.cmu.edu!bb3.andrew.cmu.edu!news.sei.cmu.edu!cis.ohio-state.edu!magnus.acs.ohio-state.edu!math.ohio-state.edu!howland.reston.ans.net!ix.netcom.com!netcom.com!jqb
From: jqb@netcom.com (Jim Balter)
Subject: Re: RACE and IQ
Message-ID: <jqbCynvHK.73L@netcom.com>
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
References: <38anh5$oho@agate.berkeley.edu> <38ffta$43s@nntp.Stanford.EDU> <38ipno$231@agate.berkeley.edu>
Date: Wed, 2 Nov 1994 22:22:32 GMT
Lines: 32

In article <38ipno$231@agate.berkeley.edu>,
Gerardo Browne <jerrybro@uclink2.berkeley.edu> wrote:
>rherardi@leland.Stanford.EDU wrote:
>
>: It seems rather important to have a clear and precise formulation of 
>: "intelligence."
>
>I disagree with this sentiment, because as I see it, discovering an
>"intelligent" entity (a person, usually) involves recognizing that
>someone exhibits familiar features, in particular, behavioral
>features that we value.  The epithet  "stupidity" is often used 
>to refer to those behaviors that we ourselves don't understand or
>empathize with.  "Religion is stupid", is a common sentiment among
>the nonreligious.  Arguments denying the stupidity of some practice
>must *justify* that practice and in doing so appeal to the
>prejudices, or shall we say values, of those who are to be convinced.
>
>It would be less misleading to say that AI tries to model familiar
>"intelligences", such as those of a human or a dog or a fly, than
>to say that it tries to build "intelligence as such".  When
>AI finally succeeds, the question to ask about a machine will be,
>"what intelligence is it like", or, if it is quite new, a description
>will have to be given.

It seems to me that "intelligence" must be related to efficacy.  A "new
intelligence" that is not effective as a tool for reaching previously set
goals should be called something else.



-- 
<J Q B>
