Solving the simulation problem

Local to this post, I will define solving to mean "figuring out what to do in the face of."

The simulation problem goes like this: given that what we do to understand, predict, and control ourselves is build simulations, that's what any other sufficiently advanced cognitive species would do to, so there will be uncountable billions of simulations of our world, so the odds that we are living in the original are miniscule.  I believe that this is what Nick Bostrom was most famous for when he got to set up the Future of Humanity Institute at Oxford.


One of two cases holds; we assume it is impossible to discriminate which.

I.  We are the original.  

In that case, the world's largest economy is an autocracy; it's third largest economy is dismantling its democracy in favour of autocracy, and the transnational organisation including the second largest economy (the euro zone) is in danger of losing one of its two member states that has a functional military.  Much of this is due to problems stemming from unsustainable resource consumption and of unequal wealth distribution.  We should worry about that.

II.  We are in a simulation.  

In that case, at least one of three cases holds, here which is a normative question.

  1. Nothing we do matters, in which case we should follow the policy set by I. however insignificant the chance I. is the case that holds.
  2. It makes no difference that we are in a simulation, our duties are to our own friends and phylogeny whether it is intrinsically biological or simulated, in which case we should follow the policy set by I.
  3. What we do matters to whoever or whatever simulated us; any solution we find will benefit them as well.  In this case, we may as well follow the policy set by I., since it has no chance of harming us and it may benefit others.

Therefore in all cases we should focus on solving the problems presented to us by our own world.