From: Anders Sandberg (asa@nada.kth.se)
Date: Fri Jun 28 2002 - 07:52:10 MDT
On Tue, Jun 25, 2002 at 09:07:52PM -0400, ABlainey@aol.com wrote:
> After the singularity, I feel it would be fair to say that we will
> still have crime. In fact I feel it fair to say that as long as WE survive in
> some form or another we will still have crime.
Yes. But the consequences of crime and the incidence of crime could
well be different. If people have backup copies murder of a body
would be closer to assault and vandalism than "real" murder; if
anybody can replicate most material posessions theft might become
more akin to software piracy or stealing post-it notes. Also, in a
society dominated by very smart agents rational selfishness would
be a strong factor - and even if criminality is part of human
nature, so is altruism and a sense of justice. They would likely be
similarly expanded in scope.
The real issue is the problem of destruction, that the destructive
power of individuals is growing. This means that malign or
irrational individuals can do more damage.
Let's see, today it is demonstrably possible for one or a few
people to kill a few thousand people in an act of terrorism. A
hundred years ago I expect that the most one could do would be to
go rampage with a machine-gun, which would likely kill a few tens
of people before being stopped. Around 1800 the weapons were even
cruder, and the numbers would be lower. At medieval technology you
would have to use slow to reload crossbows or whack people with
swords; I guess a medieval killing spree would be around ten
people. If we compare this with the growth of population, we have
had a roughly sixfold increase in population but maybe a
hundredfold in destructive power. The risk is that this destructive
power continues to increase to such an extend that individuals
become threats to all people.
Solutions suggested to this have been either along the line of
relinquishing advanced stuff (not terribly likely, and not
evolutionarily stable), leaving it to an enlightened elite that
knows what to do with them (but who to trust?), leaving it to some
automatic trustworthy system (the old nanarchy or the more modern
sysop idea; again, why trust this system?), making people nice
(either by assuming niceness memes to spread or by direct
modification), spreading out (a good idea if it can be done in
time) or instituting fairly detailled monitoring under the control
of many (the transparent society).
By having a multipolar society with strong accountability and
openness, power abuse becomes harder and more powerful
crime-prevention systems can be used to deal with real risks (a
sysop or enlightened elite is a system with the risk for a single
point failure; a network of watchers that watch each other and the
world is requires far more unlikely failures/corruptions to fail).
Mental redesign could be ethically iffy (is society for the
individual or the individual for society?) and might not even work
on certain mental architectures. Spreading out is probably the
single most efficient way of reducing risks, and it is hard to
achieve under a centralized protection agency.
I think the goal ought to be a highly resilient system that can
bounce back from any likely attack, not necessarily unhurt but
with the ability to continue to grow and flourish. We don't want to
embed our post-singularity societies in a brittle matrix.
-- ----------------------------------------------------------------------- Anders Sandberg Towards Ascension! asa@nada.kth.se http://www.nada.kth.se/~asa/ GCS/M/S/O d++ -p+ c++++ !l u+ e++ m++ s+/+ n--- h+/* f+ g+ w++ t+ r+ !y
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 09:15:04 MST