Re: [sl4] Friendly AIs vs Friendly Humans

From: natasha@natasha.cc
Date: Mon Oct 31 2011 - 13:52:38 MDT


   Quoting Philip Goetz philgoetz@gmail.com[1]:

>> From what I've overheard, one of the biggest difficulties with FAI is
>> that there are a wide variety of possible forms of AI, making it
>> difficult to determine what it would take to ensure Friendliness for
>> any potential AI design.
>
> There are 4 chief difficulties with FAI; and the one that is
> most important and most difficult is the one that people
> in the SIAI say is not a problem.

   It seems that friendly may not be, and probably is not, the
characteristic/behavior most consequential to AI agency and/or
personhood.  For example, people are often friendly, but do not have
conscience.  An empathic AI would be friendly if and when that
behavior is warranted, but not as a default (i.e. phony behavior or
spurious characteristic). 

   Natasha

Links:
------
[1] mailto:philgoetz@gmail.com



This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:01:05 MDT