Do we live in a simulated universe, Part 2: Rejoinder to Nick Bostrom.

Someone pointed out to me on a philosophy list that I hadn’t read the original article that postulates the theory that we are living in a simulated universe. He was right. I was responding to the BBC’s synopsis of the article and predominantly other stances, such as Elon Musk’s. How to deal with it. Admit it, read the paper, and come back. Here’s what I learned.

My stance in the last article still stands, but I now need to apply it specifically to Bostrom’s arguments. The first thing that struck me is he makes precisely the same argument I made a decade ago about time travel. And while his theory is clouded in big talk, to make it worthy of attention, it can be summed up so pretty much anyone can understand it inside of 5 minutes (he says 30).

My time machine argument was as follows. We will never be able to travel in time because if we ever would, we would have come back already. It’s a really simple argument. Bostrom’s is virtually the same, but he tacked on a couple other possibilities.

But first we need to know what he’s talking about, and this wasn’t really clear in the other material I looked at. We already know what time travel is. Bostrom is talking about an artificial universe populated by sentient artificial beings. It’s important to note that creators of this universe are NOT aliens, or AI, but rather future humans, creating an “ancestor simulation”. It is “Westworld”. If we are in a simulation, we are just like the “host” robots in both the original or the new “Westworld”. If you haven’t seen either, there’s a theme park people can visit where they can act out their fantasies of living in the past, such as in the old West. The old town they’ll visit is populated by robots, there for our entertainment purposes, which can be as lascivious as one likes, or as violent, or both. In the old version something goes wrong and the robots rebel, but the newer version is closer to Bostrom’s model because the robots becomes conscious, and then rebel. According to Bostrom, we could be just like those robots in Westworld, but we are in a non-physical, mathematical simulation. Got it.

Here are Bostrom’s main contentions. I like to simplify things in my own words to make them digestible, so my “translations” follow.

(1) the human species is very likely to go extinct before reaching a “posthuman” stage;

1) We will die off before we figure out how to make a simulated universe with conscious subjects in it (Westworld) .

(2) any posthuman civilization is extremely unlikely to run a significant number of simulations of their evolutionary history (or variations thereof);

2) We develop the technology but don’t use it (because it would be Westworld, and immoral…)

(3) we are almost certainly living in a computer simulation. It follows that the belief that there is a significant chance that we will one day become posthumans who run ancestor-simulations is false, unless we are currently living in a simulation.

3) We live in a computer simulation (Westworld) because if we don’t, that means we will never develop the technology to make a computer simulation (Westworld).

[Another possibility is we will develop the technology, but we aren’t in a simulation created in the future. This is similar to saying we will develop a time machine, but we won’t go back to before 2016. But there are already enough silly arguments to contend with.]

His final argument is similar to if I had declared a decade ago that time travel must have already taken place, because otherwise it would mean we would never develop the technology. It’s a pretty weak argument. There’s no evidence that time travel has occurred, so it doesn’t make sense to declare that it must have occurred anyway, but we couldn’t detect it, because otherwise we are saying it’s not possible. In short, the argument goes: it must be possible, because the alternative is it is not possible. The easy answer to this is #1, which is the same conclusion I came to about time travel: we will never develop the technology.

The reason we won’t develop the technology is that it’s more than just creating a thoroughly convincing virtual world, like a super advanced video game. The real difficult part is creating the intelligent, conscious lives to put in the virtual world. They would have to be what I call, “conscious avatars” or even more to the point “living avatars”. We’d have to create a Duke Nukem who is actually alive, and as intelligent as we are.

Now why would I say that’s impossible? Because all living things we know of are physical and organic. A robot, no matter how smart, is not alive. I do love the idea of artificial intelligence coming alive, and I’ve made art about it.

The main problem is not intelligence, but being self-aware. The computer who beat the reigning Chess champion (Deep Blue Vs. Garry Kasparov) didn’t even know it was playing chess, or that it was plugged in. We’d have to first create non-biological life, then purely mathematical life, which is non-physical. And we are talking about creating beings like us. So, while we need large, physical, biological brains to be self-aware and intelligent, we would have to recreate this in an avatar without a brain.

It may simply be impossible to create non-biological life, and/or non-physical life. That’s the most likely answer, because we aren’t capable of doing either of those things at all to any degree. Robots that look and act like humans are no more alive than Big Blue. Unique cellular life we’ve created isn’t intelligent, and is biological.

We are talking about creating “Christine”, the car that comes to life in the movies.

The strongest conclusion was not mentioned by Bostrom. Let’s call it #4.

4) We don’t live in a simulated universe because it’s impossible (to create non-biological AND non-physical life).

I think the possibility he didn’t mention is the most likely one. So, there are three compelling reasons why we don’t live in a simulated universe.

1. It’s impossible to create non-physical, non-biological life (let alone that it is also sentient, and highly intelligent)
2. Even if it is theoretically possible we will never achieve it.
3. There is no evidence that we are in a simulated reality, and there is overwhelming evidence we are biological beings in a physical universe.

~ Ends