This post refers to the video clip below. His problem with the simulation hypothesis seems to be that if you accept that you are in a simulation you can make the same argument that you are in a simulation of a simulation for the same reason. This is a reductio ad absurdum because you could go on and on forever. This doesn’t seem true at all to me. You could have simulations within simulations but wouldn’t the amount of computational resources to do this expand geometrically? You could argue that you are likely in a simulation of a simulation but it would be orders of magnitude less likely that you are in a simulation of a simulation of a simulation simply because simulating a universe is not computationally free despite the fact that we have the resources, it appears, to simulate a vast number of them.
He also argues that if we’re in a simulation we don’t know the real laws of physics. This is correct but if you start from a position that we are in a simulation you have already conceded. As I see it there are two options:
1. We could be in a real universe that is capable of simulating trillions of universes.
2. We could be in a simulated universe. In this case Tegmark is correct we don’t know the real laws of physics. The “real universe” could be, for example, tiny and only capable of simulating one other universe. However, you cannot disprove the likelihood of this being a simulation due to the possibility that the odds are more even in the real universe without conceding the argument.
As I see it the logic goes if 1 likely simulated if 2 100% simulated. The only way we can not know the laws of physics is if this is a simulation and its bizarre to use that premise to somehow refute the simulation hypothesis.