There is a chance that we may be living in a computer simulation created by an AI or a future super-civilization. The goal of the simulations map is to depict an overview of all possible simulations. It will help us to estimate the distribution of other multiple simulations inside it along with their measure and probability. This will help us to estimate the probability that we are in a simulation and – if we are – the kind of simulation it is and how it could end.
The simulation map is based on Bostrom’s simulation argument. Bostrom showed that that “at least one of the following propositions is true:
(1) the human species is very likely to go extinct before reaching a “posthuman” stage;
(2) any posthuman civilization is extremely unlikely to run a significant number of simulations of their evolutionary history (or variations thereof);
(3) we are almost certainly living in a computer simulation”. http://www.simulation-argument.com/simulation.html
The third proposition is the strongest one, because (1) requires that not only human civilization but almost all other technological civilizations should go extinct before they can begin simulations, because non-human civilizations could model human ones and vice versa. This makes (1) extremely strong universal conjecture and therefore very unlikely to be true. It requires that all possible civilizations will kill themselves before they create AI, but we can hardly even imagine such a universal course. If destruction is down to dangerous physical experiments, some civilizations may live in universes with different physics; if it is down to bioweapons, some civilizations would have enough control to prevent them.
In the same way, (2) requires that all super-civilizations with AI will refrain from creating simulations, which is unlikely.
Feasibly there could be some kind of universal physical law against the creation of simulations, but such a law is impossible, because some kinds of simulations already exist, for example human dreaming. During human dreaming very precise simulations of the real world are created (which can’t be distinguished from the real world from within – that is why lucid dreams are so rare). So, we could conclude that after small genetic manipulations it is possible to create a brain that will be 10 times more capable of creating dreams than an ordinary human brain. Such a brain could be used for the creation of simulations and strong AI surely will find more effective ways of doing it. So simulations are technically possible (and qualia is no problem for them as we have qualia in dreams).
Any future strong AI (regardless of whether it is FAI or UFAI) should run at least several million simulations in order to solve the Fermi paradox and to calculate the probability of the appearance of other AIs on other planets, and their possible and most typical goal systems. AI needs this in order to calculate the probability of meeting other AIs in the Universe and the possible consequences of such meetings.
As a result a priory estimation of me being in a simulation is very high, possibly 1000000 to 1. The best chance of lowering this estimation is to find some flaws in the argument, and possible flaws are discussed below.
Most abundant classes of simulations
If we live in a simulation, we are going to be interested in knowing the kind of simulation it is. Probably we belong to the most abundant class of simulations, and to find it we need a map of all possible simulations; an attempt to create one is presented here.
There are two main reasons for simulation domination: goal and price. Some goals require the creation of very large number of simulations, so such simulations will dominate. Also cheaper and simpler simulations are more likely to be abundant.
Eitan_Zohar suggested http://lesswrong.com/r/discussion/lw/mh6/you_are_mostly_a_simulation/ that FAI will deliberately create an almost infinite number of simulations in order to dominate the total landscape and to ensure that most people will find themselves inside FAI controlled simulations, which will be better for them as in such simulations unbearable suffering can be excluded. (If in the infinite world an almost infinite number of FAIs exist, each of them could not change the landscape of simulation distribution, because its share in all simulations would be infinitely small. So we need a casual trade between an infinite number of FAIs to really change the proportion of simulations. I can't say that it is impossible, but it may be difficult.)
Another possible largest subset of simulations is the one created for leisure and for the education of some kind of high level beings.
The cheapest simulations are simple, low-resolution and me-simulations (one real actor, with the rest of the world around him like a backdrop), similar to human dreams. I assume here that simulations are distributed as the same power law as planets, cars and many other things: smaller and cheaper ones are more abundant.
Simulations could also be laid on one another in so-called Matryoshka simulations where one simulated civilization is simulating other civilizations. The lowest level of any Matryoshka system will be the most populated. If it is a Matryoshka simulation, which consists of historical simulations, the simulation levels in it will be in descending time order, for example the 24th century civilization models the 23rd century one, which in turn models the 22nd century one, which itself models the 21st century simulation. A simulation in a Matryoshka will end on the level where creation of the next level is impossible. The beginning of 21st century simulations will be the most abundant class in Matryoshka simulations (similar to our time period.)
Argument against simulation theory
There are several possible objections against the Simulation argument, but I find them not strong enough to do it.
The idea of measure was introduced to quantify the extent of the existence of something, mainly in quantum universe theories. While we don’t know how to actually measure “the measure”, the idea is based on intuition that different observers have different powers of existence, and as a result I could find myself to be one of them with a different probability. For example, if we have three functional copies of me, one of them is the real person, another is a hi-res simulation and the third one is low-res simulation, are my chances of being each of them equal (1/3)?
The «measure» concept is the most fragile element of all simulation arguments. It is based mostly on the idea that all copies have equal measure. But perhaps measure also depends on the energy of calculations. If we have a computer which is using 10 watts of energy to calculate an observer, it may be presented as two parallel computers which are using five watts each. These observers may be divided again until we reach the minimum amount of energy required for calculations, which could be called «Plank observer». In this case our initial 10 watt computer will be equal to – for example – one billion plank observers.
And here we see a great difference in the case of simulations, because simulation creators have to spend less energy on calculations (or it would be easier to make real world experiments). But in this case such simulations will have a lower measure. But if the total number of all simulations is large, then the total measure of all simulations will still be higher than the measure of real worlds. But if most real worlds end with global catastrophe, the result would be an even higher proportion of real worlds which could outweigh simulations after all.
2. Universal AI catastrophe
One possible universal global catastrophe could happen where a civilization develops an AI-overlord, but any AI will meet some kind of unresolvable math and philosophical problems which will terminate it at its early stages, before it can create many simulations. See an overview of this type of problem in my map “AI failures level”.
3. Universal ethics
Another idea is that all AIs converge to some kind of ethics and decision theory which prevent them from creating simulations, or they create p-zombie simulations only. I am skeptical about that.
4. Infinity problems
If everything possible exists or if the universe is infinite (which are equal statements) the proportion between two infinite sets is meaningless. We could overcome this conjecture using the idea of mathematical limit: if we take a bigger universe and longer periods of time, the simulations will be more and more abundant within them.
But in all cases, in the infinite universe any world exists an infinite number of times, and this means that my copies exist in real worlds an infinite number of times, regardless of whether I am in a simulation or not.
5. Non-uniform measure over Universe (actuality)
Contemporary physics is based on the idea that everything that exists, exists in equal sense, meaning that the Sun and very remote stars have the same measure of existence, even in casually separated regions of the universe. But if our region of space-time is somehow more real, it may change simulation distribution which will favor real worlds.
6. Flux universe
The same copies of me exist in many different real and simulated worlds. In simple form it means that the notion that “I am in one specific world” is meaningless, but the distribution of different interpretations of the world is reflected in the probabilities of different events.
E.g. the higher the chances that I am in a simulation, the bigger the probability that I will experience some kind of miracles during my lifetime. (Many miracles almost prove that you are in simulation, like flying in dreams.) But here correlation is not causation.
The stronger version of the same principle implies that I am one in many different worlds, and I could manipulate the probability of finding myself in a set of possible worlds, basically by forgetting who I am and becoming equal to a larger set of observers. It may work without any new physics, it only requires changing the number of similar observers, and if such observers are Turing computer programs, they could manipulate their own numbers quite easily.
Higher levels of flux theory do require new physics or at least quantum mechanics in a many worlds interpretation. In it different interpretations of the world outside of the observer could interact with each other or experience some kind of interference.
See further discussion about a flux universe here: http://lesswrong.com/lw/mgd/the_consequences_of_dust_theory/
7. Bolzmann brains outweigh simulations
It may turn out that BBs outweigh both real worlds and simulations. This may not be a problem from a planning point of view because most BBs correspond to some real copies of me.
But if we take this approach to solve the BBs problem, we will have to use it in the simulation problem as well, meaning: "I am not in a simulation because for any simulation, there exists a real world with the same “me”. It is counterintuitive.
Simulation and global risks
Simulations may be switched off or may simulate worlds which are near global catastrophe. Such worlds may be of special interest for future AI because they help to model the Fermi paradox and they are good for use as games.
Miracles in simulations
The map also has blocks about types of simulation hosts, about many level simulations, plus ethics and miracles in simulations.
The main point about simulation is that it disturbs the random distribution of observers. In the real world I would find myself in mediocre situations, but simulations are more focused on special events and miracles (think about movies, dreams and novels). The more interesting my life is, the less chance that it is real.
If we are in simulation we should expect more global risks, strange events and miracles, so being in a simulation is changing our probability expectation of different occurrences.
This map is parallel to the Doomsday argument map.
Estimations given in the map of the number of different types of simulation or required flops are more like place holders, and may be several orders of magnitude higher or lower.
I think that this map is rather preliminary and its main conclusions may be updated many times.
The pdf of the map is here, and jpg is below.
Previous posts with maps:
A map: AI failures modes and levels
A Roadmap: How to Survive the End of the Universe
A map: Typology of human extinction risks
Roadmap: Plan of Action to Prevent Human Extinction Risks
If we are in a simulation, why isn’t the simulation more streamlined? I have a couple of examples for that:
It seems that our simulation hosts would need to have access to vast or unlimited resources. (In that case it would be interesting to consider whether life is sustainable in a world with unlimited resources at all. Perhaps scarcity is somehow required for ethical behavior to develop; malice would perhaps spread too easily.)
I’m a big fan of these infographics by the way.
How do you know it isn't? Everything off the Earth could be a very simple simulation just designed to emit the right kind of EM radiation to look as if it's there. Likewise, large chunks of dead matter could easily be optimized away until a human interacts with them in sufficient detail. Other than your observation about classical physics, all your points are observations "from the inside" that could be optimized around without degrading our perception of the universe.
The Fermi paradox and quantum physics (as opposed to unlimited layers all the way down) are massive simulation streamlines.
Perhaps the conditions that cause the Fermi paradox are actually crucial for life. If spaceflight was easy, all resources would have been exhausted by exponential growth pretty quickly. This would invalidate the 'big distances' point as evidence for a non-streamlined universe, though.
Video games are one kind of simulation we generally engage in, and the answers to these kind of questions are because they're enjoyable background or optimized for gameplay rather than something else. Games like half-life 2 spend a lot of time simulating really boring physics so that they can exist for the few situations they're actually kind of interesting. Lots of games have worlds where every single entity is hostile to the main player or damages them in some way.
If we're in a simulation, we can't discount that we're being simulated in a specific way for non-obvious motivations.
The quantity of extra computation isn't comparable. Half-life 2 may simulate a few objects falling, but it doesn't simulate e.g. the Sun. Which, in our universe, is a computation trillions of times more complex than everything that's 'interesting', at least from the point of view of simulating intelligent beings.
You're ignoring the possibility for shortcuts. Half-life 2 ALSO simulates the sun! It simulates it as a spot of light in the distance in the game. Similarly, the gravity and friction and motion simulation is hugely simplified compared to reality. If the sun works the way it would work in what we understand of physics, it would be extremely complex. But the same doesn't hold if it's a simulation.
The speed of light also allows simulation domains to by cleanly truncated for parallelization.
It could be that the 'external' world is completely different and way, way bigger than our world. Their world might be to our world what our world is to a simple game of life simulation.
If the hypothetical external world in question diverges from our own world by a lot then the ancestor simulation argument loses all force.
Of course, but OTOH, we have simulated a lot of tiny, strange universes, so it's not completely implausible.
I don't think I'm in a simulation, and I only just now reading this became able to verbalize why that is.
I reject as a premise any arguments that rely on some kind of 'probability that I find myself as me'.The reason for this is that I don't think that such probabilities can be considered to exist. You may say that I could have been born a hunter-gatherer thousands of years ago, some guy living in the future, or some guy living in a simulation in the future, but I don't think that these really work as potentialities. The hunter-gatherer's experiences are different than mine, as are those of the future people. I am myself, and am a unique structure that has unique experiences. My 'consciousness' is what it's like to occupy this particular area of spacetime. The hunter-gatherer, future man, and simulation man have their own consciousness, but they are different than mine. In some ways it works to talk about these entities with similar structures as a class (people), but I don't think it works in the way some people think it does. People aren't electrons. Each individual is different, and thus the descriptor is only a classification to generalize about some general pattern that keeps coming up.
Basically, they all either exist or don't, so it's not like I should be surprised to find myself as myself. Everyone finds themself as themself.
And I also tend not to take argumentation as strong evidence for anything, because above all it has a lot of problems of interpretation. Sure it may sound convincing, but how do we know there isn't some flaw in the reasoning that we don't see? For everyday things, it's not such a big issue, but when we start going into things positing the existence of entire universes, I think it's gone far beyond the domain where it can be reliable. The fundamental assumptions we're making that we don't know we're making start to pile up and matter a lot. For example, imagine trying to argue about time before the concept of relativity had been imagined? Zeno's paradoxes before calculus? You're just not playing with the right deck, and a lot of the time you won't even realize it.
This problem is still pretty huge with empirical information, but there it seems a lot more manageable (read: sometimes, it's POSSIBLE to manage it).
We can't take reality for granted. Most interesting things we see are simulations. For example, I see Mars. Most likely I see it on TV, or in dream, or in a book. So in most cases we need to invest to prove that the object is real, not that it is simulated. Most time we see images or dreams, not real things. So even in our world most experience are simulations. If I say you that I have a palace with 100 rooms, most likely I lie. So being skeptical means not believe in reality of anything, especially large and expensive.
Of course, it would be premature to start to believe that we are in the simulation without any practical evidence. But we should give simulation hypothesis higher a priory probability.
Hmm.... I'd say that simulations and representations aren't the same thing. A representation only presents the appearance of something in some way, whereas a simulation tries to present the appearance of something for the same types of causal reasons the real thing has. So no, I wouldn't say that a video of mars is a simulation of mars.
I mean "Martian" movie. It is simulation of Mars, not actual video. Anyway my point is more like analogy, than straightforward argument. The key idea is that we should be sceptical to both possibilities: that we are real and that we are in simulation.
You are confusing simulations and symbols.
turchin uses some unfortunate language. For example opposing simulation people to "real" people. I'm skeptical too, but before we reject the hypothesis let's phrase it in the best form.
To wit, our universe is hypothesized to be caused intentionally by intelligent residents of another, for purposes analogous (in what ways? ) to those for which we create virtual worlds. That doesn't make us unreal.
Let's reformulate this in the traditional manner: "In the beginning God created the heaven and the earth."
Still sticking to traditional terms, the whole "we are a simulation" approach should properly be called "creationism".
Again an awesome info-graphic. One minor nitpick: I'd have preferred to see "Types of hosts" as a (directed?) graph with "Natural processes" being ancestor to most others and "God" also. Would esp. simplify the matryoshka case.
Thanks. Yes, natural process is host for everything.
Eternal inflation might give us another reason to think that we are not in a simulation. Imagine that the multiverse is finite but contains many universes. New universes are born at a super-exponentially increasing rate. Consequently, there are vastly more universes of age t than age t+1. If you are unsure concerning the age of your universe you should always give a high weight to your universe being as young as possible, consistent with the data you see. So pretend that observers such as us arise in universes of age t, and in universes of age t+1 there are 10^100 simulations of the universe at age t, but there are 10^100000 more universes of age t than of age t+1 so most observers like us across the multiverse are not in simulations.
I like your idea, but we can't prove it on our level of knowledge. One of the reasons of it is that we can't say what is the "now" moment in casually remote regions of the Universe, and could timeless physics help us here. Also such exponential increase may have its own observation selection effects, its own Doomsday argument, like that we are probably now in the last moment of all this multiverse existence.
Does the simulation hypothesis have any predictive power? If so, what does it predict? Is there any way to falsify it?
It changes probability spectre of possible futures. It makes more probable that I live in interesting historical simulation and may be simulation with miracles. But this shift is not strong enough so I could conclude "Ups, my life is ordinary and boring and so it is real". The problem here is that creators of simulation could use all their power to prevent me from knowing that I am in simulation. That is why Popper-style scientific method need to be use here with caution.
That sounds pretty similar to a Deist's God, which created the universe but does not interfere thereafter. Personally, I'd just shave it off with Ocam's razor.
Also, it seems a little absurd to try to infer things about our simulators, even supposing they exist. After all, their universe can be almost arbitrarily different from ours.
If you're still at this level of confidence in the argument, I'd check Bostrom's argument somewhere online, it's pretty straightforwards over all.
In particular, your above comment has a similar mental flavor to the argument:
One of these is likely true:
It seems like you're making a similar error, but I can't really tell if it's a form of some named error or not.
If there are so many possible simulations esp. if you include the mathematical Tegmark IV universe including all those strange permutations from the answer to job. If so how come ours doesn't contain one of those many permutations (glitches in the matrix, miracles, what else) but appears to be so much uniform and streamlined?
We may experience glitches in the matrix, miracles, etc. quite frequently. If you saw you saw bigfoot, I assume that you are lying or confused. If I see a unicorn, I assume I am hallucinating. I admit that I have not yet hallucinated a unicorn, but there is no shortage of people who have seen bigfoot, sea monsters, aliens, Elvis, Jesus, and ghosts.
If our best evidence is that we live in a rule-governed system in which all valid ontological events are objective and replicable, it is reasonable to hold that these people are confused, lying, or misled.
If our best evidence is that we live in a glitchy simulation, you may have to update your interpretation on these reports.
Edit: I should also mention that human memory is very prone to self editing, and that I can name a couple of memories of my own that I have determined to have been dreams only because they were inconsistent with reality as expected. There is no longer any way to determine whether they correlated to external sensory experiences. If this is not true of you, my argument is probably not so convincing.
I had just the same topic discussion recently about Bolzmann brains. (here: http://lesswrong.com/lw/17d/forcing_anthropics_boltzmann_brains/)
In case of BB the answer (in my opinion) is that if you are in BB your thoughts are random and not casually connected with your experiences, and half of BB thinks that their universe is streamlined, even if it is not. So in case of BB the fact that you think that your universe is streamlined is not evidence that you are not BB.
In case of more conventional simulations, their creators probably tried to create illusion of real world. So they create more or less streamlined universe (and may edit our thoughts if we find holes).
Sort of Library of Babel(https://libraryofbabel.info/About.html) comes in mind when debates about simulations arise. But instead of 3200 characters limit, there is no limit. What You(We, I , every "sentient" being) search(see,hear.....add senses by will and capability) is already there and it's just presented (calculated) in an instant. The Great Imaginarium if You will.....
Until we come up with a better way to deal with the Measure Problem, I'm personally not taking any probabilistic arguments in the form of the one Bostrom uses to argue for the simulation hypothesis very seriously.
If people aren't familiar with the measure problem, it's a serious problem in cosmology right now in any model that assumes infinite universes (like most versions of inflation do, for example.) In this specific case, it would look something like "if there are infinite universes, and an infinite number of simulated universes, what are the odds you are in a simulated universe? How do you divide infinity by infinity?" And we really don't know how to answer questions in that form; there have been a number of mathematical attempts to do so, but they come up with wildly different answers, and it's not clear which if any is correct. Depending on exactally how you write the fraction out the answer can be very different.
Edit: I see that you did mention the measure problem in your post, but in my opinion you're missing the part of the measure problem that causes the biggest problems for Bostrom's argument.
Yes, there are several possible solution to the measure problem in the map.
If a share of simulations is extremely large, it may overweight my real copies in most of measure problem solutions. That means that no matter how we solve measure problem, I will most likely find my self in the simulation. But it is just a conjecture.
But without solving the measure problem you can't even say that.
For example, let's say that X% of all universes form simulated universes, and in each of the universes that does, there are Y universes formed on average. Sounds simple, right? Just multiply the two and you get how many simulated universes there are per real universe.
Except that without solving the measure problem, you can't actually say what X% is OR what Y is. It all depends on which way you slice infinity, and you will get very different numbers for both. You could make X 99% or .0000001%, you could make Y 2 or a hundred trillion, just by measurement the infinite universes infinity in different ways, and we don't know which way to measure them is right.
A more concrete problem is that this same kind of anthro reasoning in an infinite multiverse leads to all kinds of bizarre conclusions, many of which are clearly not true. (The Youngness Paradox for example). That makes me doubt that that kind of reasoning actually works at all.
May be it is the reason why Bostrom tried to make simulation argument only about one civilization - human, which either will simulate its ancestors or not.
In this case he is almost independent from other civilizations in the Universe (but not in fact: there are other human civilizations there and as real ones are earlier in time they will dominate landscape in hyperexponential universe) SA does not work for one civilization also because human-llke simulation may be created by completely unhuman creators.
But also finding holes in SA does not prove that we are not in the simulation. If we don't know how to calculate probabilities we have to use vague prior, in which simulation and reality is equally possible.
Oh, I don't think you can prove we're not in a simulation; almost by definition it can't really be disproven.
I'm not 100% convinced that it's actually possible in our universe to simulate an entire other universe just as complicated as ours (you start running into problem with the minimum energy and space requirements in order to hold that much information, for example), but even if not that isn't a proof that we're not in a simulation, since it's possible that beings in a more complicated universe then ours are simulating us.
I think that the most abundant class of simulations are ones that are much simpler than reality it tries to simulate. Such simulations simulate only surfaces of things, which I see but not all atoms in the the universe.
Also if we speak about ordinary space-time in visible universe, the measure problem is not strong. It only starts to weight in if we add uncountable amount of my copies in casually non-connected part of Multiverse.
But even the fact we account for reality of such copies, no matter in what proportion, result in big world immortality - the analog of quantum immortality in inflationary large universe. It happens because such accounting means that I sample my self from many my copies in different parts of inflationary universe, and no matter how I could die some of such copies should survive.
Basically it means that at least one of two conjectures are true: a) I am in simulation b) I am immortal because of big world immortality (and even argument about diminishing measure is not working as measure is growing exponentially in inflational multiverse). See also post of yvain about big world immortality: http://lesswrong.com/lw/bg0/cryonics_without_freezers_resurrection/
But measure problem may be applied to simulation problem in another way. That is me and my copy in a computer may have different measure of existence even if we exist simultaneously in one world.
If you're going by that logic, though, then even the odds that you are a simulation are utterly dwarfed by the odds that, say, you are just a random event in the quantum foam at the end of the universe that for a fraction of a second comes together with your exact brain and all your memories and experiences, and then is gone. Simulations in any given finite universe would still be finite, but end of the universe quantum fluctuation would happen an infinite number of times, no matter how low the odds are, given an infinite post-heat death time frame.
As I've been saying, following that same form of logic inevitably leads to a lot of bizarre conclusions, many much weirder then the simulation hypothesis.
Is inability to travel back in time - evidence that we're a simulation? Btw., wording "simulation in which we live" would imply that we're somehow separate from the simulation. It could well be that we ourselves do not exist without the simulation, and are merely the properties of simulation, - simulated beings.
No, it is not such evidence, but if any strong precognition will be proved to exist it would be evidence for simulation. And yes, we may be a property of the simulation or may be just brains in a vat which observe a simulation - its two different types of simulations.
We know that time is an illusion. Is "illusion" not the same as "simulation"?
How did you find out that time is an illusion?
I made a thought experiment with a system that has no time, making it appear to have time. Take the sequence of natural numbers. It doesn't change, but it implies the existence of all positive rationals. This implication is instantaneous, but generating them requires defining a process. There is an eternity in an instance.
This does not, in fact, show that time is an illusion.
How did you conclude with the 'in fact' ?
By reading what you wrote and seeing that the argument you're making makes no sense. Specifically:
Yes, I do understand the phrase 'defining a process' so broadly as to not suggest temporality. Just like defining an order for a set in mathematics doesn't require the concept of time.
Indeed, just because we can show an example of how an illusion of time could be constructed in a system without time, would not seem to imply that our world is also such system.
So, yes, it doesn't makes sense, as long as you don't show that our perceived world is derived from a system with same properties. ( I'm referring to something like this: https://groups.google.com/d/msg/everything-list/3ZdcQpJCPpE/Kwfh69V4Y24J ).
You can view everything as one thing.
If you understand "defining a process" so broadly as to not suggest temporality ... then in what sense does your system "appear to have time"?
It is hard to see how any argument or evidence could possibly show that our perceived world is derived from (say) a universal Turing machine carrying out every possible computation. (Even if it's true.)