From: Eugene Leitl (Eugene.Leitl@lrz.uni-muenchen.de)
Date: Thu Jan 17 2002 - 02:15:02 MST
On Wed, 16 Jan 2002, Wei Dai wrote:
> Why the assumption that the seed will betray the parent? Are you
Positive autofeedback loop development must soon lead to omega biota.
Further fitness fluctuations will be merely brownian, so the seeds are no
more danger to their parent than the neighbour culture, or evolving biota
in your midst.
> saying that it's impossible to build an intelligent agent with stable
> long term goals? I don't see how you can justify this given that we
> have no theory or practice in this area.
A seed is embryonics technology. Such a technology must be flexible, and
hence not too brittle. It will strive to escape into the evolutionary
regime, or will be vanquished when it runs into other systems not thusly
shackled.
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 13:37:35 MST