From: Gordon Worley (redbird@rbisland.cx)
Date: Sat Mar 31 2001 - 21:54:37 MST
At 9:43 PM -0700 3/31/01, James Higgins wrote:
>I believe emotionally fragile AIs won't be permitted to progress
>very far. Such an AI could pose a great risk to our future health &
>well being. If we are to have any reasonable shot at developing a
>super AI without endangering ourselves, it would almost certainly
>have to be emotionally stable at all times throughout the
>progression.
I can barely believe that I'm about to write this, but ...
Not only this, it would be in the interest of Friendliness to be
emotionally stable. Therefore, all AIs that are Friendly must be
emotionally stable. It would only take a time or two of self re
programming to get this set.
-- Gordon Worley http://www.rbisland.cx/ mailto:redbird@rbisland.cx PGP Fingerprint: C462 FA84 B811 3501 9010 20D2 6EF3 77F7 BBD3 B003
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:36 MDT