Re: A conversation on Friendliness

From: Damien Broderick (thespike@satx.rr.com)
Date: Thu May 27 2004 - 10:29:02 MDT


As an exercise, examine the semantic content of Eliezer's interlocuter's
contributions:

At 04:39 AM 5/27/2004 -0400, EY wrote:

>I recently had a conversation on FAI theory, which I am posting with
>permission for the sake of illustrating that FAI theory doesn't always
>have to be painful, if both parties follow something like the same
>communication protocol:
>
> <James> Nothing terribly communicable. I am wondering if a correct
> implementation of initial state is even generally decidable.
> <James> I'm having a hard time seeing a way that one can make an
> implementation that is provably safe.
> <James> At least generally.
> <James> There might be a shortcut for narrow cases.
> <James> right
> <James> but how do you prove that the invariant constrains expression
> correctly in all cases?
> <James> sorry. stays friendly
> <James> at runtime
> <James> Right
> <James> precisely.
> <James> that's the problem
> <James> heh
> <James> I've been studying it, from a kind of theoretical implementation
> standpoint. Very ugly problem
> <James> No thoughts yet.
> <James> Yep. Always have to be aware of that
> <James> I've noticed. Seems like a reasonable approach
> <James> Don't know if it is optimal though
> <James> for whatever "optimal" means
> <James> I'm not satisfied that I have a proper grip on the problem yet
> <James> Too many things I define lazily

How is this distinguishable from the responses of an Eliza-bot?

Damien Broderick



This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:47 MDT