META: Dangers of Superintelligence

From: Emil Gilliam (emil@emilgilliam.com)
Date: Sat Aug 28 2004 - 19:34:38 MDT


On Aug 28, 2004, at 8:19 PM, fatherjohn@club-corsica.com wrote:

> One safeguard that might be tried is to make sure that any AI has a
> very limited connection to the "real world" perhaps just one computer
> screen and a keyboard. That way the computer won't just grow out of
> control.
>

I wonder why nobody ever thought of that before? Oh, wait ...

    http://yudkowsky.net/essays/aibox.html
    http://sl4.org/bin/wiki.pl?AI_Jail

Please review the SL4 archives, in at least enough detail to know which
subjects have already been argued to death. These pages may be useful:

    http://sl4.org/bin/wiki.pl?DeadHorse
    http://sl4.org/bin/wiki.pl?CategorizedSL4Archive

- Emil



This archive was generated by hypermail 2.1.5 : Tue Feb 21 2006 - 04:22:43 MST