Simulation Dilemma - Human Existential Crisis, Understanding the Past and Future Theory of the Whole Universe

Elon Musk often has some outlandish ideas, and one of his most outlandish ideas is that we are likely to live in a simulated environment. In interviews and press conferences, he gave a playful logic to explain the idea, and everyone was hooked! But is he right?

This puzzling logic was first proposed in 2003 by Nick Bostrom, who named it the simulation hypothesis. It goes like this:

The development of computers has grown exponentially, and the computing power level of future civilizations will greatly exceed the current computing power. We already simulate many aspects of life through computers, from video games to art to scientific research. These future civilizations may simulate the universe from its origin to its end, understanding the universe and humanity in precise detail. There are billions of simulations in the simulation, all of which carry human history. So, the odds that you are your original self and not one of many simulated yourself are one in a billion. From this, we can conclude that we are most likely simulating.

Pretty persuasive logic, isn't it? It sounds like we have no choice but to accept that we are in a virtual reality world. But it doesn't, because this is a logical argument with no defined parameters or statistics. Musk is doing the same thing as quantum healers and other pseudoscientists, creating an idea that sounds airtight and convincing, but hasn't been tested in reality. I can hear Newton rolling in the grave!

But with a little math, some testable concepts, and a little critical thinking, we can unravel the logic of this existential crisis and answer the question: Are we living in a simulated environment?

What are the arguments for Musk's claim?

There is only one way one can make a strong argument for a simulation hypothesis, and that is by discovering the shortcuts programmers take when developing our simulations.

The first is based on rendering and the speed of light. We've all played computer games, and there's fog in the game. This is likely to allow the computer to only render the scene close to the player character. If the entire scene had to be rendered, that would cause lag and stutter. Overlords who simulate humans might do something similar to ours, so our universe has a universal speed limit, the speed of light.

Due to the very large distance (4.2 light-years) between the solar system and the nearest star, it is nearly impossible for us to reach these distant objects quickly. So instead of simulating the 93 billion light-year-wide universe, Lord programmers just have to simulate our little star system and add a "background" that looks like we're in this vast universe. We can even detect this shortcut. As human exploration of the universe deepens, we start to simulate more and more universes, which means we can compete with computer rendering speed. So imagine sending a probe to Proxima Centauri only to find it's a polygonal blob, which would be very convincing evidence that we're living in a simulation.

The second argument comes from quantum physics.

In quantum physics, the universe is undefined unless you have an observer. When Schrödinger's cat is locked in a box, it has no observer, and because of that, it is in a superposition of dead and alive, because its state is undefined. Does this sound like fog from a video game? Our universe is only "rendered" when we observe it - a valuable shortcut for using less computing power in simulations.

But there's also a debate about simulation theory, which revolves around how time interacts with quantum gravity. The theory of quantum gravity is very complex, but a paper shows that it can solve many of our quantum gravity problems if time is viewed from outside our universe. This is because the observer can "fix" time on the quantum scale, which would allow general relativity and quantum mechanics to work together. If true, it would pave the way for the creation of the "Theory of Everything."

Is anyone watching the cosmic clock?

But think carefully. An outside observer of our universe, doesn't that sound like a programmer, or a computer running a task manager?

By now, you may have noticed that these arguments are actually unverifiable. These are speculations, interesting speculations that don't lead to any solid scientific results. Are the arguments against the simulation hypothesis stronger?

Before we look at the arguments against it, we need to understand some basic things about this hypothesis. This is a trilemma, not a dilemma. It has three possible outcomes to choose from, first, the computer cannot simulate the universe; second, the computer can simulate the universe, but the creatures that inhabit it have no real consciousness; finally, the simulation will exist, and the creatures in it will also be real to live. Musk ignored the first two options, so let's take a look at them.

We can't yet say whether these virtual humans are conscious, alive, or just a string of ones and zeros. Science has yet to find a sense of being alive, so arguments against simulation theory are invalid (at least in science).

Science doesn't yet understand what it's like to be alive

Let's assume that computers are getting more and more powerful, so eventually, we can do these simulations, and the life in them really exists. Can we still have these billions of simulations?

There are two important issues here.

First of all, you may not be able to get a mock in a mock. Just like Russian dolls can only be so small, the simulation can only be compressed so far. The computing power of the computer hosting the original simulation is limited, so each subsequent simulation will have less and less computing power, until you get to the point where one simulation is impossible. So it's not a one in a billion chance of being ontology, but one in five.

One more question, we don't have a unified theory of physics, so what are we simulating? Currently, physics has some important pillars, such as quantum mechanics, the Standard Model, and general relativity, but they are not well-connected. This means that physicists don't yet know what the fundamental truth of the universe is. Furthermore, we know that current theories are not enough to explain everything, just a little bit about dark matter and dark energy. So, the programmers who designed these simulations don't yet have the "basic code of the universe" to replicate, and what's worse, our best "theories of everything" like string theory, if correct, are so complex that it would take thousands of years to solve open. We have to face the fact that we may never know the fundamental truth of the universe, and if so, then these simulations cannot happen.

There is no mathematical model that can simulate everything

What about outside observers of quantum gravity? If this is the answer, who was the outside observer of the original universe? Unfortunately, this idea does not stand up to scrutiny.

So it's quite possible that even with sufficiently powerful computers and truly living virtual creatures, we still won't be able to achieve what Musk said is a one-in-a-billion chance, not even a fictional simulation.

Does such a computer really exist?

Scientists have proven that you cannot simulate the entire universe with a computer. But we don't want to simulate all reality, we just want to simulate humans.

For future civilizations to use past simulations, they would have to run faster than time. But computing the computing power required by this computer is very complex. Let's calculate the computing power required to simulate nearly 8 billion human brains in real time, which can give us a sense of scale. If this is easy to do, then these simulations might happen.

The human brain operates at about 10^20 operations per second, which is about 4.5 trillion times faster than the Intel i7 chip, and we have a population of nearly 8 billion. That means it would take about 36 trillion Intel i7s to simulate everyone in real time.

You might say that Moore's Law means that it won't be long before our computers are up to simulations. But as the switches in these chips got smaller, they ran into a problem, quantum physics preventing them from going indefinitely. The gap from one side of the switch to the other is so small that electrons can pass through by means of "quantum tunneling," de-randomizing the switch. This destroys the logic of the chip and limits processing power.

You will need to dedicate an entire planet of energy, space and resources to run a basic simulation. This is just simulating our minds in real time, not building worlds or speeding up time, and this simulation does nothing for the civilization that runs it.

We want to run this simulation faster than real time to observe and understand how humans and the universe are developing. This means that we can run thousands or even millions of times faster than the real world, which means we need a computer with thousands or even millions of times more computing power than our "basic simulation".

So even a futuristic, powerful, planet-sized computer might still not be able to host a useful simulation. Even if it could, the humans in it probably wouldn't experience conscious life, and the simulation wouldn't spawn billions of them in it.

From all of this, you can see that the puzzle isn't as clear-cut as Musk makes it sound, making some amazing assumptions and jumping to conclusions of his own. But with the advent of technologies like quantum computing, these machines, no matter how impractical or massive, may one day be built, provided we humans survive long enough to build them. But that's another topic.

But if you look at the details with a critical eye, you have to assume a ton of near-impossible things to come to the conclusion that we live in a simulation.