From: Stuart Armstrong (firstname.lastname@example.org)
Date: Mon Jan 05 2009 - 03:30:20 MST
>> the potential "Friendly AI"
> The correct term is Slave AI.
Though I'm fine with either term, it's irrelevant to the discussion.
Your objection seems to be
1) FAI as generally imagined is utterly unlikely
2) So Rolf's gamebit must fail.
But this thread was not about the possibility of FAI. The question
here is, given the assumption that FAI is possible and somewhat
likely, should rolf's gamebit be implemented; and should the tuba be
It's called a hypothetical situation, btw. Don't be afraid of it, it won't hurt.
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:01:03 MDT