Re: Why would AI want to be friendly?

From: Eugene Leitl (eugene.leitl@lrz.uni-muenchen.de)
Date: Fri Sep 08 2000 - 02:32:22 MDT


Jason Joel Thompson writes:
 
> > > Does truly superior intelligence require free will?
> >
> > Can you tell random from pseudorandom, without further knowledge about
> > the generation source?
>
> This is an interesting question.
>
> My immediate temptation was to answer 'no.'

Then the question about free will vs. determinism is meaningless in
practical terms. Even if we're clockwork puppets deep down below, we
can't use that fact to predict anything.
 
> However if a number generator is pseudo-random, then it applies is a logic
> or process by which it generates its numbers, which in theory I imagine
> might be detectable given a sufficiently large sample and sufficiently smart
> pattern recognition software.
 
Well, a good pseudorandom number generator passes all statistical
tests for randomness, and has only one pattern: it's periode, after
which it begins repeat itself. This periode can be very long,
i.e. much longer than the universe exists. Hence there is no
measurement procedure allowing you to tell the output of one from
another.

> One could assign a probability of being truly random to such a string of
> numbers. (Which is all one could do in any case.)

I'm not sure I understand what you mean.



This archive was generated by hypermail 2b29 : Mon Oct 02 2000 - 17:37:31 MDT