Re: [sl4] A model of RSI

From: Matt Mahoney (
Date: Fri Sep 26 2008 - 15:30:45 MDT

--- On Fri, 9/26/08, Stuart Armstrong <> wrote:

> > If winning an election is good, then becoming supreme
> dictator of the earth is better. If running a successful
> company is good, then acquiring all of the world's
> wealth and starving the rest of the population is better.
> I never said that they would be safe tests. They are tests
> you should only run if the AI is friendly to start with.

So we need to test for friendliness *before* we test for intelligence. How do you propose we do that?

-- Matt Mahoney,

This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:01:03 MDT