[extropy-chat] The Simulation Argument (was: Atheists launchinquisition...)

[Hal Finney]

John Clark writes:
> The argument fails if the Many Worlds interpretation of Quantum Mechanics is
> true and there are an infinite number of non simulated beings. Even if each
> real world runs 99 trillion simulated universes there are still the same
> number of simulated and non simulated worlds.

It's an interesting question, but I don't think it would work this
way.  After all, the MWI does not destroy ordinary considerations of
probability.  We still find that some events are more probable than
others, even though the MWI suggests that there are an infinite number
of worlds which experience each outcome.

This can be handled by the notion of "measure", which is a sort of tag
attached to an MWI world and measures how much probability amplitude
it contributes.  Worlds of small measure are unlikely to be experienced.

If a world of large measure runs a simulated mind, then that mind is
also of large measure.  If such worlds run a great many simulated minds,
then the total measure of the simulated minds is greater than the total
measure of the non-simulated ones, even though there are an infinite
number of each.  In this way, the same kind of reasoning applied in the
vanilla SA goes through in the MWI.

> I also think a super civilization might have less enthusiasm for running
> such simulations than some think for ethical reasons, a concept they must
> have some knowledge of if they have not destroyed themselves.

I agree that there are ethical questions involved in running simulations
of minds which don't know they are simulated.  Maybe it would be OK if
you could somehow get their permission ahead of time, along with their
agreement to forget that they would be in a simulation.  Even that seems
a little questionable; is the person who gave consent really the same
as the person who experiences the simulation?

Hal