From: Keith Henson (email@example.com)
Date: Sat Jun 12 2004 - 22:56:53 MDT
At 06:36 PM 12/06/04 -0700, Michael Roy Ames wrote:
>CV would appear to be a useful theory, and if realized as a *process* would
>provide useful data on how humanity might make decisions, but it doesn't
>sound in the least bit friendly. I suggest that if it did turn out to
>approximate "human friendliness" better than a very friendly human would,
>then it would be an accident. I do not expect humanity's collective
>volition to be friendly. Do you expect it to be friendly? If so, why?
I don't even expect CV to be consistent over time.
Does anyone doubt that wars are a result of "collective volition"?
I make an argument rooted in evolutionary psychology and evolution that the
volition leading groups into wars is a result of privation, especially
"looming" privation where after a run up the economy goes sour for either
the whole population or some sub group.
(Put "xenophobic memes" in Google if you don't remember the argument.)
In times of plenty the "gain" on xenophobic memes is low, so tribes can do
business with each other and swap women. But when times get bad . . .
humans have had to be their own predators for a long enough time for
conditional gene based psychological mechanisms to evolve.
If you are going to even consider "collective volition" and don't want
wars, then you are going to have to rewire these "go to war when times
look bad" mechanisms out of humans or keep them out of bad economic
I have no idea of what side effect rewiring would cause. It might be a
very bad idea.
At the root of these mechanisms is the tendency for human populations to
grow beyond their ecological limits. In the west, productivity has stayed
ahead of population growth for such a long time that few are aware of
it. But the mechanism is resurgent whenever there is an economic contraction.
Since the privation that turns on "collective volition" to wars is
ultimately driven by population growth, an AI that was trying to keep away
from this mode would have to strongly limit births--to very low levels for
If humans were given a choice between wars and population restrictions
I wonder what the choice would be? No doubt it would vary with cultures.
PS. With respect to the dangers of AI, friendly or not, I no longer think
anyone on this group has a handle on the problem. I am not absolutely sure
there even is a problem, but if there is I would say we are all some
distance from understanding it.
This archive was generated by hypermail 2.1.5 : Fri May 24 2013 - 04:00:35 MDT