--- John Ku <[EMAIL PROTECTED]> wrote:

> On 2/16/08, Matt Mahoney <[EMAIL PROTECTED]> wrote:
> > The result will not be pretty.  The best definition (not solution) of
> > friendliness is probably CEV ( http://www.singinst.org/upload/CEV.html )
> which
> > can be summarized as "our wish if we knew more, thought faster, were more
> the
> > people we wished we were, had grown up farther together".  What would you
> wish
> > for if your brain was not constrained by the hardwired beliefs and goals
> that
> > you were born with and you knew that your consciousness did not exist? 
> What
> > would you wish for if you could reprogram your own goals?  The logical
> answer
> > is that it doesn't matter.  The pleasure of a thousand permanent orgasms
> is
> > just a matter of changing a few lines of code, and you go into a
> degenerate
> > state where learning ceases.
> 
> Your counterfactuals seem very different from Eliezer's and less
> relevant to what matters. I think Eliezer's definition was plausible
> because it approximated the standards we use to deliberate about our
> values. As such, it is getting at deeper values or procedures that we
> implicitly presuppose in any serious discussion of values at all. Even
> if you were to question whether you should use that standard, your
> cognitive architecture would still have to do so by reference to some
> internal standard in order to even count as a meaningful type of
> questioning and Eliezer's definition would probably be a decent
> intuitive characterization of it. Of course, you are free to pose any
> type of bizarre counterfactual you want, but I don't see how
> evaluating it would be relevant to what matters in the way that
> Eliezer's would.

I admit I am oversimplifying Eliezer's definition.  Reading the full document,
we should not assume that an AGI would be stupid enough to grant our
extrapolated wish to be put in a blissful, degenerate state.  Nevertheless I
am mistrustful.  I am most troubled that CEV does not have a concise
description.

> I would prefer to leave behind these counterfactuals altogether and
> try to use information theory and control theory to achieve a precise
> understanding of what it is for something to be the standard(s) in
> terms of which we are able to deliberate. Since our normative concepts
> (e.g. should, reason, ought, etc) are fundamentally about guiding our
> attitudes through deliberation, I think they can then be analyzed in
> terms of what those deliberative standards prescribe.

I agree.  I prefer the approach of predicting what we *will* do as opposed to
what we *ought* to do.  It makes no sense to talk about a right or wrong
approach when our concepts of right and wrong are programmable.


-- Matt Mahoney, [EMAIL PROTECTED]

-------------------------------------------
singularity
Archives: http://www.listbox.com/member/archive/11983/=now
RSS Feed: http://www.listbox.com/member/archive/rss/11983/
Modify Your Subscription: 
http://www.listbox.com/member/?member_id=4007604&id_secret=96140713-a54b2b
Powered by Listbox: http://www.listbox.com

Reply via email to