I recently listened to an episode of Joe Rogan where Nick Bostrom tried to explain the concept of simulation theory. Although the progenitor of the concept tried multiple times, it seemed as though there was some roadblock to understanding, so I’d like to give it a shot. The best method would be to do it through the Socratic method, where questions are asked to allow the other participants to reach the conclusion desired themselves. That’s kinda hard through text, so I’ll just discuss the path I see as required to understand the statement below.
This short article is an attempt to explain what the following sentence means:
If the simulation hypothesis is correct, the reality we are in is most probably a simulation.
This is a profound statement that follows directly from a set of assumptions. I think Nick Bostrom failed at establishing what those assumptions are. So here we go. I’m going to state the assumptions and the framework it makes, then show how that inevitably leads to the previous statement.
Ok cool, so if we are to look at the simulation argument, then we are to believe there are 3 possible options on the evolution of intelligent existence across the universe:
- All civilizations of intelligence fail to reach sufficient advancement to get past the required technology to create simulations.
- Some civilizations of intelligence reach sufficient advancement to get to the required technology to create simulations, but for some reason, choose not to pursue creating simulations
- Some civilizations of intelligence reach sufficient advancements to get to the required technology to create simulations and continue on to create said simulations.
That’s it. We have three options for the possibility of how intelligent civilizations progress. This blog post is NOT about which option is correct, though. We want to talk about probabilities ONLY IF OPTION 3 IS CORRECT. So let’s make an assumption:
Assumption #1: In the lifetime f the universe, there will exist a civilization of intelligence that reaches the required level of technology to create simulations of the universe.
We’ve cut out options 1 and 2. That’s a big assumption, but follow along throughout this thought experiment, because it will show how those that say things in this field come to the their conclusions.
Ok cool, based on Assumption #1, we’re going to pretend that at some point in the entire lifetime of the universe, some intelligence reaches the technological complexity to make simulations that rival our current experience of reality. It is so good that it is impossible to know whether what we perceive as existence is a simulation or reality. This leads to another assumption:
Assumption #2: The simulations created by the intelligence are of sufficient quality that observers inside them would be incapable of understanding that they’re in a simulation. It is indistinguishable from “reality.”
Note that we have not yet discussed OUR place in this, only that something happens sometime by someone. We’ll get to us later.
So let’s take stock of where we’re at now. Based on the stated assumptions, a civilization of intelligence comes into being at some point in the universe and creates a number of simulations that are so good that those inside it are capable of asking themselves if they’re in a simulation. We now have two groups of people that are able to ask the question “am I in a simulation:” those in a simulation, and those that are in the reality that created the simulations.
So which group are we in? What can we say from here? We as humans on Earth are a civilization of intelligence that is capable of asking “are we in a simulation?” If we follow the above assumptions, then we can only look at the distribution of possible people able to ask that question. That’s a shit load of people in simulations and a small amount of folks in the reality that created them. Since we can’t know which reality we’re in because the simulations are so good, it immediately follows that:
If one has to pick randomly against all the potential people able to ask the question “am I in a simulation”, they are most probably going to pick from the group that is in a simulation and not the group that created the simulations.
Why? Because the group in a simulation is so much larger than the group that isn’t. That’s all they are saying. If you want to believe that a sufficient simulation can exist, then it immediately follows that you are most probably in such a simulation from a sheer numbers and statistics point of view. Let’s restate everything together.
If we are to assume that some intelligence across the entire lifetime of the universe reaches enough advancement so they can create simulations that are indistinguishable to our current understanding of existence, then we are (from a statistics point of view) most probably in one of those simulations.
Note that the above statement doesn’t mean that we aren’t the actual intelligence that goes on to create simulations, just that it is very unlikely. Based on our assumptions, anyone who is capable of asking the question is incapable of knowing if they’re in a simulation because the simulation is that good, so we have no way of knowing if we’re in one.
Hope that helps, it was fun to think about how to explain this, and frustrating to listen to the 1 hour of the episode where that didn’t seem to click.