From: Gordon Worley (email@example.com)
Date: Sun Feb 09 2003 - 14:39:24 MST
On Sunday, February 9, 2003, at 02:43 PM, Michael Roy Ames wrote:
> Your phrase, "as we think of them" leads me to doubt we are talking
> about the same definition for the word 'game'. Here are a couple of
> definitions I am using:
> game n. (Form of) contest played according to rules and decided by
> skill, strength or luck. <Concise Oxford>
> game n. A competitive activity or sport in which players contend with
> each other according to a set of rules. <American Heritage>
That's exactly what I meant by "as we think of them".
> As Ben Goertzel opined, teaching an AGI to play games could be entirely
> a positive thing, as long as ve understands, "it's just a game."
This is what's dangerous. What if the seed AI plays chess, learns what
it means to compete and win, enjoys it, and then proceeds to apply
competition elsewhere. What if it then decides that it's competing
with humans for computational resources?
Worse yet, what if it learns how to be deceptive while playing games.
Think about any sufficiently complex game and you'll realize that one
of the key strategies is not letting your opponent know what you're
going to do. Perhaps the best example of this is poker, where an
ability to deceive your opponents is key to winning. Given it thinks
it's competing for resources with humans, now it is fully able to
deceive us. We want it to be Friendly, so it pretends to be Friendly
so well that we can't tell the difference. Then, when we think
everything is okay and it starts going FOOM, we discover that it had
deceived us and are prompted turned into computronium.
If you're going to say "these are just games and you should not think
about them in relation to anything" then you might as well not have any
games played at all.
> At first, the competitiveness of the games presented should (IMO) be
> very low or zero. But as the Seed AI gains understanding of a wider
> context, competitive elements can be added with confidence that they
> (and will) be explained in that wider context. Competition can be a
> learned technique that benefits all, players and non-players alike.
> Instead of "Ha! I win! I win! Now I can squash you like a bug!" we
> have, "Hoohoo! I win! Now let's apply this new technique to solving the
> protein folding problem."
There is far more to games than winning and losing. If that were all
there were we could probably have a seed AI play games, but games are
much deeper than you seem to be thinking.
-- Gordon Worley "It requires a very unusual mind http://www.rbisland.cx/ to undertake the analysis of firstname.lastname@example.org the obvious." PGP: 0xBBD3B003 --Alfred North Whitehead
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:41 MDT