Re: Building a friendly AI from a "just do what I tell you" AI

From: Joshua Fox (joshua@joshuafox.com)
Date: Tue Dec 04 2007 - 12:30:05 MST


Thanks, yes, I might not have understood, though I thought I did. Perhaps
that was evident from my more detailed comments, in which case I now
understand (*) what Eliezer meant.

Joshua

(*) Or maybe I don't!

2007/12/4, Robin Lee Powell < rlpowell@digitalkingdom.org>:
>
> You certainly understand *something* based on reading the paper, but what
> you understood may or may not be what Tim intended you to understand.
>



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