From: Patrick Crenshaw (email@example.com)
Date: Thu Nov 04 2004 - 19:57:37 MST
I'm sure this has been covered elsewhere, but since you are
quite...flexible, I thought you might have a different view now.
The question is: Are you trying to construct something with CV that is
provably Friendly, and if so, is that in a mathematical sense,
or a physical sense?
The physical sense is much easier to think about (and more likely
correct). So, supposing we had a very simple universe (only a few
particles, or perhaps a pseudouniverse with a small system of coupled
oscillators), what would a FAI want to do in this universe? How would
this differ from a non-Friendly AI? Though this type of question may
be silly for a much larger universe, it seems to me that since the
options for an AI would be so limited, humans with a sense of humor
could likely approximate what an AI would do without too much trouble.
I'm going to resist the temptation to bombard you with more questions
along these lines, so I'll just shut up now.
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:50 MDT