Where it becomes important is what comes after. If we are a simulation, chances are when we “die”, we just get re-run, either in the same simulation or get re-used in a different one. Or we could even go to “heaven” if the programmer thought that was a cool reward for his sims.
Preach it brother!
Though, to be accurate, they ripped off the special effects and styling, rather than the plotline. So I’m not sure that this comment is particularly relevant to this thread.
We are Skynet.
There’s a principal of statistics that, if you have only one data point, your safest assumption is that the data point is an average data point. It’s really bad data, but it’s the safest bet.
Like, if you pick one random person, it’s a much safer guess that they have a 100 IQ than that they have a 140 IQ. They might be 140, but given no other information, “average” is the default assumption.
That’s why we’re a bit mystified by the Fermi Paradox. Based on our development, we should assume that half of the life in the Universe is more advanced than us and half less-so. And, since we are space traveling creatures who are sending all manner of signals off into space, any life who is more advanced than us should be noticeably space faring and/or throwing major signals off in all directions. That we’re not detecting anything else is strange. Maybe it means that we’re actually an early species. Or, maybe it means that species who are detectable across space do not last long…
But, by the same token, if we’re anywhere close to recreating life in a computer, then it quickly becomes likely that most life will be computer generated.
Within the next hundred years, it’s likely that most of our social interactions will be via computer, in virtual worlds. We’ll have AI creatures going around, performing tasks, getting into trouble, and being entertaining for us to watch. If there’s one me, and I’m going to want a giant, virtual mansion, with dozens of personal servants, and that’s true for everyone, then most people in the virtual world are going to be artificial life. If you meet a person, the average person is artificial.
So, who is to say that you or I is not an AI? There’s no evidence against it or, perhaps, we’re unable to notice evidence which would go against it by design.
There certainly are wealthy people in the world, who have all manner of personal servants, and it’s plausible that they find the rest of us to be amusing critters. They could be the “players” and the rest of us the “NPCs”. Really, the only reason to discount this is Ohio. There’s no reason for vast, dull expanses of land to exist in a virtual world. Shit-shoveling probably wouldn’t be an aspect of life in a holodeck world. Though, if we are a virtual world, there’s nothing to say that we were created to be entertaining. There’s nothing to say that 95% of the most dull, tedious, and horrible parts of life were stripped out, and since we have no concept of their existence, shit-shoveling seems like a horrible thing to us, but to the Players, it’s more equivalent to brushing your teeth. The awful things of their world can only be expressed in Lovecraftian terms.
Computers offer the possibilities for vast quantities of electronic life, locations, planets, and materials. Minus any reason to believe that it’s impossible for an unknown, unknowable universe to create computers that can simulate those things (which is, by definition, impossible to deny), the number of virtual things will far exceed the number of physical. There’s already, certainly, far more virtual swords than real ones. There may be more virtual furniture than real, by this point.
It’s not a given, but it’s entirely plausible that we’re virtual. And the better our technology becomes, the higher the odds are that we’re somewhere in the middle of an onion of virtual worlds.
If you shuffle a deck of cards and deal them out in a line, the chance of dealing the exact sequence that you did deal is about 1 in 80,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000 - really unlikely, and yet you can achieve this monumentally unlikely outcome in just a couple of minutes. How would you interpret this fact?
However, if you assume that there is an average longevity to each instance of detectible life, then you must also assume that we are around the middle of our longevity cycle, which at the most would be a few thousand years if you consider the minor terraforming we did to be able to be interpreted as signs of intelligent life by a sufficiently advanced alien intelligence. So then we could assume that each detectable species only survives for 10,000 years or so and then perhaps terraforms or nukes itself literally back to the stone age.
We have no way of knowing what laws of physics apply in the “higher” universe where the sim is being performed. You’re projecting what you know into a realm where what you know might not be true any longer.
(My view is, yeah, it’s possible…but so are all the other varieties of Cartesian Doubt. Maybe it’s a dream. Maybe it’s a delusion or hallucination. Maybe demons. Maybe Genies. Maybe totally random noise that appears as a pattern. Ultimately, the hypothesis is “nonsense” because we don’t have any way to test it.)
That’s a whole heck of a lot of words to say “I don’t know how we could do it now, so we’ll *never *be able to do it.”
But this is the part that you don’t seem to get:
“Forget about starting it last Thursday - if you knew how to set the state of the universe that precisely, you wouldn’t have to do the simulation.”
I’ll keep using NMS, just because it’s probably the best example we’ve got right now in the popular fiction. In NMS, you see a “snapshot” in time. There’s no past and future, true. But even without an emergent history, you’ve still got that planet over there, and that sun over there. They didn’t “evolve” to be there, they’re strictly a function of the mathematical calculations. Even *without *billions of years of motion and physics, they’re still there.
It’s the same thing in a simulated universe. We don’t know for certain that everything started off with a Big Bang billions of years ago- we *think *that’s what happened, based on the existing data. And that data is derived from observing the universe. We don’t have another baseline to go by- unless you’ve been hiding another universe somewhere.
Processing time doesn’t matter. Hell, for all we know, the servers are distributed and are using an entire galaxy’s worth of calculation for hundreds of trillions of years to get *one second *of our simulation- and in the real universe, it wouldn’t matter, because for all we know, there *is *no heat death in “real” physics. We don’t know how the parent universe works.
Your error is in assuming that our simulation- if that’s what we are- is a perfect simulation of the “real” universe. You’re assuming that our programmers are making a mirrored copy of their own universe, but there’s no reason to suspect that. Even NMS isn’t doing a perfect simulation- they have their own limited periodic table, for example.
If you want to know run a simulation, you only run the portions of the simulation which are important to whatever question you’re trying to answer. When you make a racing sim, you don’t spend time programming in the exact way an electron orbits an atom of carbon in the rubber of the car’s tires- you are only concerned with the interaction of the tire to the road.
That could be the way our simulation runs on the “real” universe’s computer analogues. It may take trillions of years to get one second of our simulation… but from the inside, there’s no way to tell. Their universe doesn’t have to follow the same rules- or even logic- ours does.
Or exactly what **Trinopus **said, dammit.
There’s also the possibilities that:
- It’s dirt hard to travel any appreciable distance.
- Advanced transmissions technology are too well-focused and compressed, so that there’s no noticeable structure.
- The signals we’re receiving from most of space is too old for life to developed in most of it, yet.
The Fermi Paradox isn’t insanely paradoxical.
The simulation hypothesis says nothing one way or another about the Fermi paradox. Why wouldn’t we meet simulated aliens in the simulated universe if we would meet real aliens in a real universe?
Why wouldn’t we know? Back in the '50s they called computers electronic brains, and no one I remember predicted that by today we’d have far more “electronic” brains in the world than human ones. I’d not be surprised if there more AIs than people, and all over the place. I would be surprised if a lot of them were in simulated worlds.
There may be some slight glitches in the simulation next Tuesday around 4pm ET as we switch out one of the hot drives.
There are things that can be different - the value of c, the mass of an electron, Planck’s constant. But I’d doubt that you could get a consistent universe without energy conservation laws. You might as well say goddidit then. But they hypothesis, as I understand it, does not assume anything supernatural.
You misread this part of my post. I was talking about statistics in that portion. I was not linking aliens and simulation.
Why would we?
I can envision many scenarios where the virtual denizens of a virtual world/universe would not be made privy to their situation. I mentioned several in the other thread. Here’s a few more:
- Research (which I mentioned in the other thread, but it’s such a large one, I can’t not repeat it)
- Historical reconstruction - e.g., using a monitor that was sent out at FTL, to read signals from Earth, from the past, and replay them as-is.
- Entertainment (breaking the 4th wall breaks the experience)
- Law (banning people from acting as gods over created life - as it would be immoral)
- Running without oversight (the originator isn’t paying attention, has passed away, etc.)
- Structural difficulties - e.g, the quantity of material is too large for the creators to interact with most of the stuff they have created.
If your simulation level is planets and stars, then starting with a batch of planets and stars with more or less random properties is possible. But are they consistent with others? And if you look more deeply at them, do you start to see the code?
In our universe, as far as we can tell, you can go down to quarks pretty much. Initially populating a universe at that level is a bit harder.
All simulators worth anything allow you to save state and reload in case something crashes. But where exactly do you dump several universes worth of data?
Nowhere did I say that you couldn’t start later - just that if you can accurately initialize a universe to some consistent state, you have the knowledge you’d get from doing the simulation, and so wouldn’t really have to do it at all. Sure, you can run it for 13 billion years and start up a bunch of children simulations with different laws, say, but then most are inconsistent.
In other words, magic, with no conservation laws. And even if this magic universe had a simulation, simulated worlds where there were conservation laws would not be able to simulate a universe - and so the hypothesis fails.
After all, our universe has them, so by this argument the average simulated universe must have some sort of conservation law also.
Not at all. The important thing is the level of detail. If elements, say, or even proton, neutrons and electrons, were all we saw, I could believe that these were the simulation primitives of our universe. But we can see within these. Higher level simulations are coarse, and our universe is exceedingly fine.
BTW, as I noted in the other thread, mixing simulation levels is possible, and we’ve tried it, but it leads to all sorts of problems.
Rules of logic are unlikely to be dependent on the universe. When you write a simulation model, you only make it as big as it has to be. You can’t pick yourself and go to Paris in GTA, after all. If they were simulating for something specific, why does our universe have to be so big?
If you started with the Big Bang and just clocked it for 14 billion years, sure. But you probably could set initial energy to create a smaller, easier to simulate universe. This does not depend on laws of physics - except assuming that energy and time have some non-zero costs to the simulators. If their cost is 0, you are talking about God, and all bets are off.
- Excessive fascination with the work of Philip K. Dick.
The solipistic universe hypothesis is a lot more plausible than the simulated universe one. Doesn’t make a whole lot of sense, but it doesn’t require universes where energy is free.
If we are in a video game it is a pretty fucking horrific and boring one. What is wrong with his argument is that some species had to come along and be the first ones to create those video games. I’m assuming that is us.
I do agree with his views that we will probably become pets to superintelligent AI.
I’m not sure what you mean by energy being free?
I think people are assuming Musk takes this way more seriously than he does. He said that he and his brother agreed to ban the topic from their hot-tub conversations, which I think might give a clue as to the category of topics this belongs to. They probably talk about teleporters, too.
It’s a fun claim because it’s hard to disprove, despite sounding ridiculous. And because there’s nothing at stake; no one should ever change their behavior based on the belief or disbelief in it.
The guy still runs two companies focused on solving actual problems in our current universe. If he really took the argument seriously, you’d think he’d relax a little and wait until he pops up a level.
Did you just state that “no one should ever change their behavior based on the belief or disbelief in it” – and then follow up, in the next breath, by saying you’d think he’d behave differently if he took the argument seriously?
Key words are “should” and “really”. If one really believed we were in a simulation–and I mean absolute confidence in the argument–then one might well change their behavior. But no one, and clearly not Musk, takes it that seriously (nor should they). They might think it’s very likely to be true, but as soon as you’ve given up absolute confidence then there’s not much reason to change behavior.
There are also two different probability measures: first, the question of whether we’re in a simulation, given the premises; and second, the probability that the premises and argument as a whole are true. One might indeed say that if the argument works, we’re almost certainly in a simulation, but have only 50% confidence that there aren’t any unforeseen holes in the argument or premises.