Re: Military Friendly AI

From: James Higgins (jameshiggins@earthlink.net)
Date: Tue Jul 02 2002 - 09:52:51 MDT


At 06:47 PM 6/29/2002 -0400, Brian Atkins wrote:
>Higgins seems to want "hundreds or thousands of relevant experts" to agree
>that it is ok for you to "push the big red button". Are you ok with that?

I thought that was *obviously* an exaggeration. If you got a reasonable
group of people together I doubt you'd have this type of problem. Neither
Ben nor Eliezer are likely to think a .01% chance of the AI going rogue is
unreasonable (unless there turns out to be a darn good reason to think
so). The other people selected for such duty should have a similar view
(on this subject) as Eliezer & Ben. My point is simply that individuals
and probably most development teams have too limited a perspective to make
this type of decision alone. The discussions on this list should be a good
example of that.

James



This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:40 MDT