The Simulation Argument, remixed
The other day I published my Introduction to the Simulation Argument, to some interesting reactions. Although it’s a very simple argument, the mental contortions one has to jump through make it difficult for many people to grasp. So let me recap and generalize a little.
- If there are intelligent beings
- And if they create simulations
- Then simulated worlds will outnumber non-simulated worlds.
Fairly simple so far, right? Okay, let’s make it personal:
- If simulated worlds outnumber non-simulated worlds
- And if we can’t tell the difference
- Then we should assume we’re in a simulation
That’s the tough step for most people, and leaves you working back through the chain of arguments to see where things went wrong.
For most people, the part where it goes wrong is the idea that we can’t tell the difference between a simulated world, and a non-simulated world. But that’s our usage of the word “simulation” here. We’re assuming that with the right technology and skill, someone could create a simulation so real, the participants wouldn’t be able to tell the difference.
You may question whether that would even be possible. In that case, we’re back to step 2, “…And if they create simulations”.
This step hinges on intelligent entities being able and willing to create the simulations. If they die out before they figure out how to do so, then they’re not able. If they have the capability, but somehow choose to never create any simulations, then they’re not willing.
But keep in mind what we’re saying. Just two simulated worlds would make the Simulation Argument valid. Just two simulations in all of history.
If you’re saying that these simulations will never happen, you’re saying that for the next trillion years, no intelligent being will ever create a realistic simulated world.
Given how realistic our simulations are getting these days, it seems to me that if no one ever creates a fully realistic simulation, it’s because something pretty drastic happens. So…are you feeling lucky?