Newsgroups: comp.ai
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!oitnews.harvard.edu!purdue!lerc.nasa.gov!magnus.acs.ohio-state.edu!math.ohio-state.edu!howland.reston.ans.net!newsfeed.internetmci.com!news.kei.com!uhog.mit.edu!news!ml.media.mit.edu!minsky
From: minsky@ml.media.mit.edu (Marvin Minsky)
Subject: Re: "strong AI" "weak AI"
Message-ID: <1996Feb11.071855.3366@media.mit.edu>
Sender: news@media.mit.edu (USENET News System)
Organization: MIT Media Lab
References: <4fj9q6$idu@newsbf02.news.aol.com>
Date: Sun, 11 Feb 1996 07:18:55 GMT
Lines: 23

In <4fj9q6$idu@newsbf02.news.aol.com> lshauser@aol.com (L S Hauser) writes:
>Does anyone know if these terms were in use prior to Searle's 1980 "Minds,
>Brains, and Programs" (the Chinese room article)?  Or did Searle invent
>them?

I don't recall ever hearing them before that time.  Nor have I ever
heard them since, in talking with with AI researchers.

>I'm also like to hear what people understand by these terms . . .
>especially what they understand by "strong AI".

My impression from the non-technical discussions I've heard is that
Strong AI is when a machine is "really intelligent" and weak AI is
when a machine can do anything except be "really intelligent".  The
goal of the discussion is to distinguish those machines that (in
addition to solving hard problems, etc.) also possess a certain
attribute that is not only irreducible and undetectable, but not even
describable--except by expressions like "that machine really
understands (or means, or intends, or feels X".

The goal of that, it seems to me, is to show that we're not only now
unique, but are destined to remain so.  They used to do that by
invoking the soul, but now they need something more currnt.
