From newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!news-server.csri.toronto.edu!rpi!usc!wupost!uunet!psinntp!scylla!daryl Fri Jan 31 10:26:59 EST 1992
Article 3264 of comp.ai.philosophy:
Newsgroups: comp.ai.philosophy
Path: newshub.ccs.yorku.ca!ists!helios.physics.utoronto.ca!news-server.ecf!news-server.csri.toronto.edu!rpi!usc!wupost!uunet!psinntp!scylla!daryl
>From: daryl@oracorp.com
Subject: Re: Searle Agrees with Strong AI?
Message-ID: <1992Jan29.200716.27881@oracorp.com>
Organization: ORA Corporation
Date: Wed, 29 Jan 1992 20:07:16 GMT

Wayne Peterson writes:

> The problem with the turing test and all intelligence test is their
> insincerity. Maybe the burden is on the tester to find the
> intelligence rather than sit it jugdment, missing opportunities to
> discover new answers to old questions.

I don't believe that failing the Turing Test is an indication of lack
of intelligence, only that passing is an indication of the presence of
intelligence. As you point out, autistic children may very well fail
to pass the Turing Test (or any other kind of intelligence test), but
it is because of the failure of communication, rather than failure of
intelligent.

What I have been arguing is that behavior should be a sufficient, but
not necessary indication of intelligence. I agree with you that we
shouldn't necessarily give up just because the first test turns out
negative.

Daryl McCullough
ORA Corp.
Ithaca, NY




