Well, my "The Next Wave" post was intended to be humorous.  I not that much
of a comedian, so I may have weighed in too heavily on "apparently serious."
Let me apologize to the extent it was a feebly frivolous failure.

Perhaps I am wrong, but my impression is that the talk here about AGI sense
of self, AGI friendliness, and so on is quite premature.  "The Next Wave"
post was intended to illustrate this by way of even more off-the-deep-end
topics.  Heck, the last I heard we haven't been able to write an algorithm
that can beat an accomplished 9 year old at Go.

However, I think it is quite conceivable for a future AGI to learn all
non-trivial details of the universe, arriving at point where there is
effectively nothing left to learn and nothing left to do.  Why do something
if you know in advance the result?  Even random outcomes, such as the nearly
infinite number of snowflake patterns will get boring and pointless after,
say, 8,941,204,723,493,808 images.

But if we find an infinite number of parallel universes . . . ah, forget it.

Kevin Copple


-------
To unsubscribe, change your address, or temporarily deactivate your subscription, 
please go to http://v2.listbox.com/member/?[EMAIL PROTECTED]

Reply via email to