Without wanting to speak on their behalf, I think that's exactly vidarh's point. That simulation is indeed unfalsifiable and therefore mjburgess' conjecture that it's false doesn't hold.