The core of Nick Bostrom's Simulation Argument isn't just a philosophical thought experiment; it's a probabilistic model. He boils down the possibilities into a trilemma by estimating the fraction of all observers with human-like experiences that are living inside a simulation. To do this, he introduces a few key variables.

Using these variables, we can estimate the total number of simulated individuals with experiences like ours. It's the product of these three factors: $f_p f_I \bar{N}$. The number of non-simulated, biological individuals in a civilisation is normalised to 1 for this calculation. The fraction of observers who are in a simulation, $f_{sim}$, is therefore the number of simulated individuals divided by the total number of individuals (simulated + biological).

Look closely at the equation. If the term $f_p f_I \bar{N}$ is very large, the '+1' in the denominator becomes insignificant. The fraction $f_{sim}$ then approaches 1. This means we are almost certainly in a simulation.

If that term is very small (close to zero), then $f_{sim}$ also approaches zero. This implies we are not in a simulation. This outcome can happen in two ways:

1. $f_p \approx 0$: Almost no civilisations survive to reach the posthuman stage. This is the extinction proposition.
2. $f_I \approx 0$: Posthuman civilisations exist, but they have no interest in running ancestor simulations. This is the disinterest proposition.

This reasoning leads directly to Bostrom's three-part conclusion: at least one of these propositions must be true.

For this argument to work, a few assumptions are necessary. The most crucial one is Substrate Independence an idea from the philosophy of mind. It posits that consciousness is not tied to a specific biological material, like our carbon-based neurons. Instead, it's about the computational structure and processes. If a silicon-based computer can replicate the intricate patterns and functions of a brain, then it too can host a conscious mind. Without this assumption, the idea of a simulated consciousness being indistinguishable from a biological one falls apart.

The argument also relies on a subtle but powerful piece of probabilistic reasoning called the Principle of Indifference. This principle states that if you have no evidence to distinguish between several possibilities, you should assign them equal probability. In this context, you have an experience of being a human in the 21st century. There are two possibilities: you are either a biological human or a simulated one.

Bostrom argues that you have no specific information to suggest you are in the 'original' biological group rather than the potentially much larger simulated group. Therefore, you should consider yourself a random sample from the set of all observers with your experiences. If the number of simulated observers is billions of times larger than the number of biological ones, then the probability that you are one of the simulated ones becomes overwhelmingly high.