Re: [sl4] Rolf's gambit revisited

From: Matt Mahoney (matmahoney@yahoo.com)
Date: Mon Jan 05 2009 - 10:44:05 MST


--- On Sun, 1/4/09, Norman Noman <overturnedchair@gmail.com> wrote:

> > If an AI believes the universe is real or simulated, it is because it was
> > programmed that way (on purpose or by accident).
>
>
> Only to the extent that if it believes anything, it's "only because it was
> programmed to". 2+2, sky is blue, etc. If one day everything turns into
> wireframe models and the moon is replaced with a big square that says BFRAME
> DECODER LAG, that doesn't PROVE you're in a simulation, but it certainly
> INCREASES THE ODDS.

Odds are something you calculate, not something that actually exists. There is plenty of evidence that you are already in a simulation. The physics of the observable universe is finitely computable. Occam's Razor appears to work in practice.

But evolution favors animals that behave as if the world is "real", that they have control over it, and that their existence matters. If I wanted to prove to you that you are a simulation, the easiest way to do it would be to rearrange your neurons.

> If humanity's whims aren't the source of the FAI's motivations, where does
> it get them? It sounds like you're taking issue with the possibility of
> friendly AI, which is a whole different argument.

These issues do need to be addressed, or else the discussion of Rolf's gambit does seem rather pointless. If the FAI reprograms our whims so that we no longer care about our own existence, and then destroys us, that's friendly, right?

-- Matt Mahoney, matmahoney@yahoo.com



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