From: Michael Roy Ames (email@example.com)
Date: Wed May 26 2004 - 13:24:57 MDT
--- Ben Goertzel <firstname.lastname@example.org> wrote:
>> It's true that intelligence allows the making of
>> more accurate predictions. But it also enables
>> the creation of more complexly indeterminate
--- "Thomas Buckner" <email@example.com> wrote
> Exactly. The comparison I draw is with the Weather
> Channel. [snip...] You can put up three times the
> satellites and hire ten times as many meteorologists,
> all smarter than Newton, and just maybe it buys you an
> extra day of reliability. This isn't even a
> controversial issue any more. More intelligence does
> not add entropy. But it can sure get outrun by it.
These responses have missed an important part of the point, which originally
"Intelligence is the power to know more accurately and choose between
futures. When you look at it from an information-theoretical standpoint,
intelligence reduces entropy and produces information, both in internal
models relative to reality, and in reality relative to a utility function."
Ben and Thomas acknowledge the ability of intelligence to make more accurate
predictions - great, we all agree on that. The part that was missed is that
intelligence has the ability to change the future through its own action.
Intelligence, taking actions in pursuit of goals, reduces entropy from an
information-theoretical viewpoint because it constrains the universe down a
path of its choosing. The 'forecast' analogy, as presented, does not hold
because the weather system contains no intelligence. If a substantial
portion of the Earth's atmosphere was composed of a kind of Atmospheric
Utility Fog (AUF) optimized for weather control, then the 'forecast' analogy
would apply. The difference between Satellites & Meteorologists as opposed
to AUF is that AUF is intelligent and takes actions toward goals.
Furthermore, intelligent systems can learn, and become more intelligent,
thus gaining more ability to channel the future in ever narrower directions
of their choosing. This is the reason why we are all here trying to figure
out how to build AI without burning ourselves.
Michael Roy Ames
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:47 MDT