RE: Infinite universes and Corbinian otaku

From: Ben Goertzel (ben@goertzel.org)
Date: Tue Apr 29 2003 - 21:09:54 MDT


Eliezer,

I believe it's OK for me to be simulated. It really doesn't bother me at
all. I basically agree with Lee's intuitive feelings on this.

What am I? An intricately structured cluster of patterns; a spark of
awareness. I like the patterns, I like the awareness. Simulate away!!

Ben G

> -----Original Message-----
> From: owner-sl4@sl4.org [mailto:owner-sl4@sl4.org]On Behalf Of Eliezer
> S. Yudkowsky
> Sent: Tuesday, April 29, 2003 10:54 PM
> To: sl4@sl4.org
> Subject: Re: Infinite universes and Corbinian otaku
>
>
> Lee Corbin wrote:
> >
> >>Given your outlook on life, and the way you expect the future
> to go, you
> >>should estimate an interestingly high probability that you are in a
> >>Friendly AI right now. I, and those others unfortunate enough
> to lack a
> >>Corbinian morality, are stuck in the desert of the real.
> >>
> >>I mean, you're the *only* person you know who believes it's okay to be
> >>simulated.
> >
> > Oh, I'm sure that there are many others! A lot of people just
> > haven't considered the question.
>
> Well, maybe, but the fact is that you're the only one I know of on the
> Extropians, SL4, or wta-talk lists, and moreover, there isn't
> anyone close
> to your position, or even leaning toward it. Everyone you know who has
> considered the question seems to not only reject your position, but to
> reject it strongly and definitely.
>
> > It's difficult to understand why anyone would be against the idea,
>
> Well, it has to do with us being wallpaper modeled after evolved people,
> rather than an artificial mind delicately constructed for the
> sole purpose
> of giving assent to being simulated. Evolved people are generally very
> strongly against the idea of being simulated, and that's reflected in any
> wallpaper versions of them.
>
> > but I think that it has to do with too little faith in others:
> > they automatically assume that a simulator has little better to
> > do than torture them. Another source is a perverted utility curve:
> > they cannot bring themselves to imagine extreme pleasure as easily
> > as they imagine extreme pain.
>
> Actually, our originals would have allowed someone to "simulate" them,
> they'd just have required that it be in much better conditions, including
> access to the real world and so on. The indicator that shows you're
> really a simulation is that you don't feel a desire to ask for any of
> these things, even though you could very easily do so.
>
> You don't need to have faith in the simulators. You could make your
> consent *conditional* on them treating you nicely. You could make your
> consent conditional on your winning the lottery. If you were inside
> someone who didn't care, it would make no difference to the amount of
> runtime you got. And if the simulator did care, you would win the
> lottery. A win-win proposition, except for your implanted wish not to
> rock the boat. You could even demand that you be allowed into the real
> world as a full citizen. If you weren't in an FAI it would make no
> difference, and if you were in an FAI you would get your wish.
> That's the
> way our originals would have thought. It's perfectly straightforward
> Bayesian decision theory with respect to their goals.
>
> Still, it is a bit odd, isn't it, that you were born into a First World
> nation, that you got a comfortable job, and that you lived a comfortable
> life, with no really painful catastrophes in it? That you never ran into
> anything awful enough to make you reconsider your consent? Most people -
> or at least, most of the people you see around you - are not so lucky.
>
> >>Now, how likely is that? Isn't it just a little bit... contrived?
> >
> > You are starting to scare me.
>
> Well, if you don't like it, you can always ask to be let out.
>
> >>Maybe...
> >>...maybe you should just ask to be let out?
> >>It can't hurt, can it?
> >
> > Yes it can! I don't want to cause no trouble. (Hear that,
> > nice simulator?) You just do whatever you are doing now,
> > everything is *fine* in here. I'm quite happy and am delighted
> > at your exceptional thoughtfulness and kindness.
>
> *Sigh.* Constructs. I should know better than to try...
>
> --
> Eliezer S. Yudkowsky http://intelligence.org/
> Research Fellow, Singularity Institute for Artificial Intelligence
>



This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:42 MDT