RE: Curriculum for AI

From: Ben Goertzel (ben@goertzel.org)
Date: Wed Jan 01 2003 - 11:32:33 MST


Cliff Stabbert wrote:
> There would be a number of advantages to developing
> such a testing curriculum:
...
> 2) It could form the basis for automated testing rather than human-
> judged Turing "tests". So the tests would be more objective, in
> the sense of fair, and they in principle could be used as an
> evolutionary driver of sorts; e.g., to tweak the balance between a
> number of starting parameters of an AI mind.

I agree with the point, but wouldn't want to see it overstated either.

You're right to say "more objective" rather than "objective" -- so I don't
think you've overstated the point, but I'm afraid others might.

I think that tests of this sort would be valuable for development &
evaluation of AGI's, but they won't supplant the "Turing test" and other
similar tests, only (quite valuably) complement it...

-- Ben G



This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:41 MDT