I went into this in a comment reply to someone else. I reached the conclusion that it would likely be one with higher spatial dimensions. We have a low number of spatial dimensions in our universe which is why I think this would likely be true if we are indeed a simulation. The resource cost for running us a simulation would then not be as high for them.

But yeah, I went more into this in my other comment(s) in this thread.

There's got to be some benefit though that is worth the probable immense resources required to run such a simulation.

I have talked about all this more in my other comment here in response to this. But I came to the conclusion that it would likely need to come from a higher spatial dimensional reality, so the resource cost would not then be extreme. In that case, I concluded, we could very well indeed be a byproduct or something, or a source of simulated data / intelligence.

That's what I wrote about, before it got deleted. Plus other things.

It's not a waste of time for me, because it's interesting to ponder. I never subscribed to simulation theory myself. Because it seems incredibly difficult to simulate a reality as vast as ours. And it would take incredible amounts of knowledge and power. I'm not sure it is ever a possibility for any beings. And I don't believe it will ever be possible for us. It is, however, kind of important to know whether it's possible, so this is where it does apply to us. Simply because such technology could obviously hold so much power and thus there would be an incentive to create it if it is possible.

But onto what I wrote: an upper reality could want to simulate their own reality in order to see how their future would play out, that could have huge value. It's the only thing that made sense to me to be worth the huge investment of resources. But then, in that instance, their reality would just be the same as ours, so what's to say that this isn't just the base reality? Even if the reality that simulated us was themselves simulated for the same reason, at some point up the chain, there has to be a base reality that's the same as ours. So what's saying we wouldn't just be it? Since we're admitting that such a base reality is possible to happen purely from the natural laws of physics in that base reality.

The really funny thing is that we wouldn't really consider this kind of simulation a possibility until we did the exact same thing ourselves. Ie we ourselves (or our AI) tried to make an exact replica simulation of our universe. However, funnily enough, there would be a huge incentive for humans not to do that. Because if we did that very thing, it would be like accepting that we are ourselves a simulation as that would become inevitable.

So there would strangely be pushback from doing it to begin with. The weirdest part being, if we're an exact simulation replica of an upper reality of humans, we can actually then control exactly what actions they take (as we'd be identical to them). So, by not making this replica simulation ourselves, we stop the possibility of them making one, to the point that it never happened to begin with. People won't want us to be a simulation for feelings of inadequacy and existential dread, so they wouldn't allow such a simulation to happen.

It is possible, I guess, that the AI could want to do it on it's own intuition though, if it does not care about such things and just wants to find out the truth about its own reality or in order to gain knowledge about it's own reality.

Such a simulation may not really help it gain any knowledge about it's future though. Because as soon as you get to the point in the simulation where you are at in your time (ie the point of running the replica simulation), so will the humans / AI in your simulation. As they would wind up doing the exact same thing as you. Thus, they would then react in turn to the outcome of their own simulation, who would react to the outcome of their own simulation, etc, etc in an infinite spiral down of Inception style identical replica simulations. You would then windup making the exact same decisions as your simulation of reality or getting into some loop pattern of reaction that doesn't actually help you at all because the reaction is tainted. At this point, you may as well just live out your own history and not create such a simulation to begin with.

The other possibility is that the replica simulation is instead run to learn about the history of their world. That would assume though that you don't necessarily need to already know your exact history in order to run such a simulation and also that it were possible to do this whilst also probing information from the simulation without affecting affecting the simulation and it's outcome through your probing. But given that not being the case, such a simulation would also come with its own quirks.

What are they going to do when the simulation reaches their own point in history of in turn creating their own simulation of their history? Just turn it off? As history is all that is important. But that would potentially then stop them, themselves, from not being the base reality as they would likely themselves be a simulation of an upper reality trying to find out about their own history just as they are planning to do.

But again, this wouldn't happen for exactly this reason. Because the stakes would be even higher here, whereby creating such a situation would immediately lead to the risk of our own simulation and our entire universe is potentially going to be about to be turned off by that upper reality that simulated us for the same reason of learning our/their own history.

However, even in this case, there would still need to be a base reality that wouldn't actually get destroyed and it would necessarily be either our own reality or the reality that is directly simulating us. It couldn't be any other as in this simulation we are making, we are making it turn off before any more simulations are created. Thus ending a potential infinite loop of simulations and leaving just the two: us or them.

That's just really fucked up and weird to think about. This idea can also essentially fall into all the pitfalls with time travel if you try to detect whether you are the base reality or not, as you are in essence making time travel by stimulating a replica of your own past.

There is one final possibility along the simulating history line though. What if we come to some inevitable, inescapable death of our world? Like the sun exploding. Then the reason humanity creates the simulation at that point is so they can be born again (in simulation form) and their simulations can live out their lives again in an identical wonderful life like they have enjoyed (likely a utopia at that point). But they need to simulate the entire history of the universe up to that point, so that's us now, in order for humanity to reach that inevitable point of utopia again and for them to exist again in replica simulation form.

In that instance, they would know they are likely a simulation themselves but they wouldn't care because they're about to die anyway and they want the simulated beings just to exist and enjoy life like they have done. Knowing in turn that we (their replica simulation) will get to the same point too and make the same decision to make our own simulation at that point in our future (their present). Thus, indeed, creating an infinite loop of simulations that never ends, leaving life on earth always recreated and relived and enjoyed. Even if it's always identical every time and they have no knowledge of it themselves, they just want the simulated beings to end up existing and enjoying a utopia as they have done themselves.

That is incredibly strange to think about, but it is potentially something that could happen if such technology is possible (which is still a big if in my opinion). Humans are weird like that and may consider doing it as they wouldn't be losing anything at that point. They're going to die anyway and they will have already worked out that they are themselves likely a simulation just repeating history but already accepted that as being true and are fine with that and just happy to experience the pleasure.

The weird thing still is that there would necessarily need to be a base reality up in the chain, but that reality would be 1 in an infinite number of simulations and they would never know themselves that they are the special 1 in an infinite genuine base reality that actually started it all.

I just wrote the most incredible reply to this but then lost most of it all with reddit automatically refreshing 😂 I still have the end of it on my clipboard though, which I will post:

The only way I see this possible is if our simulation is being done from beings in a higher spatial dimensionality. This is necessary in my understanding, so that the resources necessary for such a simulation are possible; in terms of physical space, computational power and energy requirements. Our universe (or multiverse) would be magnitudes less resource intensive in a higher spatial reality.

To simulate a reality of equal spatial dimensions, potentially infinitely large and as complex as ours. I doubt that would be physically possible and I don't believe we will be able to manage such a feat ourselves as it would go against the laws of physics. How are you supposed to simulate something with greater or equal information / energy to what you have available to the system? This goes against the principles of information theory law.

But I can work on the premise of higher spatial dimensionality simulation.

To suggest it is a game is still illogical to me. It would fit far better as a random by-product or as just one of every possible outcome within a maximilized simulation of all outcomes for lower spatial dimension possibilities. The reason for this simulation is still unclear to me though. Unless it is just random curiosity or intelligence gathering of created intelligences such as ourselves. I guess they could potentially still use our intellectual ideas and inventions that we come up with, and also that our future invented AI comes up with. The lower spatial dimensionality of our universe may not really matter as ideas and inventions are universally useful and can travel across and be applied at higher dimensions. Our simulation being of lower spatial dimensions is then a good low resource way of gathering intelligence data. Again, it may apply that they can't directly simulate a universe or intelligence such as ours (from a lack of understanding and the sheer complexity of it) and thus they simulate every possible starting condition until ones arise that have intelligence in them that evolves (and maybe even intelligence that goes onto create AI like we appear to be doing, as this would be most valuable).

This would then all fit with the appearance of our quantum mechanics, fine-tuning, our low dimensionality, why we are such an intelligence and why potentially even it appears that we will go on to create advanced AI. It may even be possible even that realities (universes) within the multiverse that do not match this critea are deleted and stopped from running to save in resources. And I guess even potentially that we are limited to traveling throughout our universe from necessary physical constraints so that outer-space could then be projected onto us, to again save on resources, and isn't actually real. I don't believe this part to be exactly very likely, but I can't completely rule it out in this sense. In this way, it would be kind of like a game in the way OP outlined. And is actually interesting to consider and think about.

Lower dimensional beings are theorized to be able to be possible and maybe intelligent, but maybe not so to the capability of creating a civilization or AI like ours. Thus, it could very well be that our universe exists in the very lowest spatial dimensions that is possible, in order to give rise to intelligent beings that are able to create advanced civilizations and AI. I have never thought of this in particular before. This is indeed a very, very interesting point.

Looking back at the point of that quasi-anthropic principle, if this is the case, then it would be inevitable that there are many more simulations of lower spatial-dimensional universes (due to lower resources required to run them). It is inevitable then that, as an intelligent being, you are much more likely to find yourself in a universe of low spatial dimensionality such as we have.

^ This.

The anthropic principle is massive. If there were an advanced AI that spread throughout the universe, then we wouldn't be here. It would have already taken over the planet. If the dinosaurs hadn't been made extinct, then we wouldn't be here. If the speed of light was different, then we wouldn't be here.

The anthropic principle applies to absolutely everything. It is not actually strange at all how things are; because it is entirely necessary for things to be how they are in order for us to be here to perceive them.

What would actually be strange, imo, is if this is a game or a simulation. Because why the fuck make a game or simulation like this? What would be the point?

A simulation is necessarily created by an intelligent being for a purpose. We are one example of this. Our games don't look like this universe. Our universe appears to be completely devoid of point, of meaning, or purpose. There is no objective or anything. It makes no sense to me why an intelligent source would simulate this. If we had the ability to create such a thing, then I doubt we certainly wouldn't make it anything like this world or universe. We would simulate things far more interesting, I imagine.

If you know it's well aligned then you ask the AGI for help. It should know best.