On Thu, 28 Feb 2008 08:32:57 -0500, "jamesgangnc"
wrote:

All your examples are still simply the consequences of the increase in
computer performance. Chess playing computers simply leverage the machine's
much faster ability to examine potential outcomes faster. They track
previous moves by other players to "learn" what to assign the higher
probabilities to. Again, predictable logic. Chess is a game with finite
possible outcomes. Admittedly a very large number of possibilties but
finite never the less. The fact that a computer can become undefeatable in
chess is no different than your ability to never lose to a child in
tic-tac-toe. In the world of cold logical actions, machines will eventually
be superior to us in just about everything. Presently their ability to
interact in the physical world is limited but that will change rapidly.

Good point, but I would posit that a human's ability to define and
adjust is no different than a computer's ability to do the same. The
main reason that Kasparov complained was that the computer, in a fit
of creativity, did something that he didn't expect. In his view, only
a human mind could resolve the permutatiaons of his attack and respond
the way Deep Blue did. When the logic streams were eventually
published, it turns out that there wasn't any human intervention - the
computer did exactly what it did by resolving and "computing" (if you
will) the variables involved and developed a move based entirely on
the current positions and downstream by ten moves.

At this point, I would like to point out that I'm not a chess master
by any stretch of the imagination - I play, but I'm not very good at
it because I'm not a strategically oriented type of person being way
to impulsive - call it a difference between strategy and tactics. I'm
great on short term gain, if you will, but lose the overall strategic
battle - call it win the battle, lose the war.

Having said that, it's my understanding that chess masters do much the
same as the computer- they look at current possibilities, extend the
position variables to moves ten and twelve iterations out, examine the
potentials of response based on past incllinations, experiences, etc.,
and develop their attacks, defenses and traps based on those
variables.

You may not wish to define this creative move as "inventive" but
rather the result of cold logic circuitry - you are looking at it from
a literal perspective and I won't argue that. I would prefer to look
at it as creative because the results were creative and as evidence
would seem to indicate, so did Kasparov as the move took him totally
by surprise as in unexpected.

Now, you could argue that he might not have accepted the same move by
a human as "creative", but I wouldn't place a bet on it.

But there are areas of human behavior that defy logic. We don't understand
those areas. What was going on in Pacasso's mind when he produced a
masterpiece? What is love? Why do we sometimes save another human, even
our enemy, when doing so is completely against our own interests? Leaving
higher dieties out of this, something is going on in our brain that appears
to rise above computational. Is it simply some sort of computational
algorythm that we do not understand yet? If so then maybe machine
intelligence is predictable. But so far it doesn't appear to be explainable
by a computational solution.

Very good points, but I think that are missing an important
distinction.

Human emotions, or even concerns, might not necessarily be the
defining element of computational intelligence. Intelligence does not
necessarily encompass those features that make us who we are. Is it
necessary that Homo Robiticus be artists, musicians or have
compassion? I would suggest that it's not necessary to be a replicant
of us because at that point, they become us. What I am suggesting is
that artificial intelligence can be self-aware with all that implies
sans that which makes Home Sapiens what it is.

In short - a whole different life form.

But to think that the expansion of computational ability at some point
crosses some magic threshold and becomes intelligence is unsupported by any
theory. There is no reason to predict that outcome and every reason to
think that it will simply be whole lot of computational ability. And a
whole lot of computational ability does not mean one of them will
spontaneously pick up a paint brush or a pencil and create something. Or
decide to kill us.

I suspect we will eventually agree to disagree. :)

And as this is WAY outside the pervue of a, in theory, boating
oriented newsgroup, I yield to you for the last word.

It's been fun.