Are You Living In a Computer Simulation?

by Nick Bostrom

This paper argues 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. 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. A number of other consequences of this result are also discussed.

Genres: Philosophy, Science, Nonfiction, Essays, Physics, Technology, Artificial Intelligence

14 pages, ebook

First published January 1, 2003

About the author

Nick Bostrom is Professor at Oxford University, where he is the founding Director of the Future of Humanity Institute. He also directs the Strategic Artificial Intelligence Research Center. He is the author of some 200 publications, including Anthropic Bias (Routledge, 2002), Global Catastrophic Risks (ed., OUP, 2008), Human Enhancement (ed., OUP, 2009), and Superintelligence: Paths, Dangers, Strategies (OUP, 2014), a New York Times bestseller.

Bostrom holds bachelor degrees in artificial intelligence, philosophy, mathematics and logic followed by master’s degrees in philosophy, physics and computational neuroscience. In 2000, he was awarded a PhD in Philosophy from the London School of Economics.
He is recipient of a Eugene R. Gannon Award (one person selected annually worldwide from the fields of philosophy, mathematics, the arts and other humanities, and the natural sciences). He has been listed on Foreign Policy's Top 100 Global Thinkers list twice; and he was included on Prospect magazine's World Thinkers list, the youngest person in the top 15 from all fields and the highest-ranked analytic philosopher. His writings have been translated into 24 languages. There have been more than 100 translations and reprints of his works. During his time in London, Bostrom also did some turns on London’s stand-up comedy circuit.

Nick is best known for his work on existential risk, the anthropic principle, human enhancement ethics, the simulation argument, artificial intelligence risks, the reversal test, and practical implications of consequentialism. The bestseller Superintelligence, and FHI’s work on AI, has changed the global conversation on the future of machine intelligence, helping to stimulate the emergence of a new field of technical research on scalable AI control.

More: https://nickbostrom.com

Reviews

[tiago. · Rating 5 out of 5]

Mindblowing is the word that first occurs to me to describe this article. It really can change your perspective on the universe; at least, it did change mine when I first heard about the simulation argument, years ago. Somehow, I only came around to reading it now, and I can confirm it is every bit as brilliant as expected. Not the easiest of reads - lots of technical lingo, some equations thrown around, a few references to ideas such as Bertrand's Paradox or the Turing test, that might require some light research - but perseverance rewards, and in few places is this statement truer than in this paper.

Bostrom analyses the possibility that a given civilization (namely one such as the human civilization) will be able to produce a simulation which integrates which conscious beings. What Bostrom proposes is that one of the following must be true:

1. The fraction of human-level civilizations that reach a posthuman stage (that is, one capable of running high-fidelity ancestor simulations) is very close to zero.
2. The fraction of posthuman civilizations that are interested in running simulations of their evolutionary history, or variations thereof, is very close to zero.
3. The fraction of all people with our kind of experiences that are living in a simulation is very close to one.

He explains this with more detail, seeking to explain how each of these hypotheses may come to being, but the third hypothesis certainly presents a disruption of the ways we traditionally see our universe. It is possible that any human-level civilization can render itself extinct before thet are able to produce high-fidelity simulations, and in our particular case we certainly have been further from that point (hypothesis nr. 1). He further points there is also the possibility, albeit a slim one, that we choose not to create such simulations, for ethical reasons for example (hypothesis nr. 2).

But accepting that both of these hypothesis are wrong, he says, then certainly, a simulation shall be produced. More than that: it can be expected that such a civilization will produce not one, but many simulations; there is further the possibility that there might be more than one such technologically advanced civilization, meaning that there will be several civilizations producing such simulations; and furthermore, the consciousnesses that inhabit those simulations may themselves seek to produce a simulation within a simulation, and that new simulation can in turn seek the same objective; and so on, and so forth. What results from this is that the number of simulated consciousnesses is vastly superior to that of non-simulated ones. Therefore, it is vastly more probable that you are within a simulation than without. In other words, in all likelihood, we are all inside the Matrix.

[Martin Krauskopf · Rating 5 out of 5]

For some reasons, as a layman in the topics, I thought that the essay would be, to some extend, easy reading - only until I had immersed into the paper. As one gets into the core of the document argumentation gets gradually more complex. Furthermore terms like "substrate-independence", "weak indifference principle", "functionalism", "computationalism", "sub-neuronal level", "principle of indifference", "Doomsday argument", etc. start to pop up. The essay explains some terms to a certain level directly, while for others reader is pointed to the bibliography or have to research in parallel with reading. The content also presumes a basic knowledge of probability theory. Even putting all the possible above barriers aside, the content is still dense to process per se, a delightful brain-twister I would say.

As such, it is a mandatory, blow minding reading for futurist, philosophers, technological geeks and other such species, or just to a non-specialists who started to be interested in involved topics

Above is subjective experience and for some, it might be easier reading. Obviously, depends on reader's background and intellect. Taking notes during the reading helped me a lot. As well as going through in parallel, or first, some related resources like the simpler explanation by author or this article, which are less dense and use less technical terminology could help.

After all, this was joyful, demanding reading and I'm sure I'll reread it one day, analyse it furthermore and look up some critiques and advances since paper gets published in 2003.

By the way, the author seems to be deservedly in an enviable situation :)

I am in a very fortunate position. I have no teaching duties. I am supported by a staff of assistants and brilliant research fellows. There are virtually no restrictions on what I can work on. I must try very hard to be worthy of this privilege and to cast some light on matters that matter.

-- http://www.nickbostrom.com/ (worth to click if you enjoyed the paper)

[Mike · Rating 4 out of 5]

It is no wonder Musk and others reference this work. The logic is sound and it seems pretty convincing to believe that the safest bet is not only to believe you yourself are not real, but there is actually a God afterall, although he probably answers to his God...and it's turtles all the way down.

[Keith · Rating 3 out of 5]

In my opinion, the key takeaway here is in the "if." Specifically, "if" (and, practically speaking, only if) such simulations are indeed not only possible, but possible and likely to be run in significant numbers, then and only then does it perhaps become useful to conclude that we are almost certainly living in such a simulation. The odds seem low to me.

Also, as has been noted elsewhere, "Simulation Theory is just Creationism for atheists."

[Dan'S_mind]

extension://elhekieabhbkpmcefcoobjddigjcaadp/https://www.simulation-argument.com/s...

20 years old perdictions still

[Snowdenwald · Rating 5 out of 5]

Maybe but probably not. My Cr(SIM)≈0 because I think that fp≈0. The assumption of substrate-independence seems to do most of the work here. Very enjoyable argumentation nonetheless.

[Forked Radish · Rating 4 out of 5]

"On the Surface of the Earth, macroscopic objects in inhabited areas may need to be continuously simulated, but microscopic phenomena could likely be filled in ad hoc."
It's certainly brilliant and insightful, however:
The first flaw in this treatise is the idea that this is a simulation per se viz. a replica or copy of something else, but this world is impossible in any context other than an "artificially" or "naturally" generated 2D MMORPG with a 3D simulation. The second flaw is that "future humans" are responsible for this world, but this is nothing but anthro-egoism. The answer IMO is that this is a "real"* 2D world with a 3D simulation imposed by necessity and teleologically, and that human technical/cultural advances and this system's modus operandi may be merely coincidental. Anything beyond this is mere speculation, interpretation, opinion, and/or conjecture. For example: humans and beavers have similar dam building technology, does this mean that humans created beavers of vice versa? Ergo, the confluence of this world's technology and human technology may not be an indicator of any cause and effect relationship, but mere mutual necessity.
*As real as anything can get. Isn't an MMORPG like the WoW just as real as your very own butt that you're sitting on right now?

[catherine ♡ · Rating 4 out of 5]

Outlandish and yet believable.

[kezzie ʚ♡ɞ · Rating 4 out of 5]

✩ 4.5 𝓼𝓽𝓪𝓻𝓼 *ੈ✩‧₊˚༺☆༻*ੈ✩‧₊˚
~
[Read for my Philosophy 101 Class]

[Rohit Jain · Rating 5 out of 5]

During World War 2, Konrad Zuse, a German mathematician working for Nazi government, after inventing the first electronically programmable computer Z3, put forth the concept of Digital Physics. This infant field of physics had then postulated that the entire universe is nothing but a computer and it follows a certain algorithm.

This premise has gained wide acceptance today, and Nick Bostrom's paper is a further exploration of this idea.

Whilst reading this article, I came upon this brilliant idea, which was directly impacted by something I was researching for years. We all know that according to ancient Indian religious belief's, Lord Mahavira and Gautam Buddha had attained liberation and enlightenment respectively viz. liberated from the cosmic cycles of birth and death. What if in reality they had found bugs (Software Developer's terminology for loopholes/ backdoors in a code) in the computer program that our ancestors have designed. What if they realized that our current gods are in reality super sophisticated programmers from the future?

For the first time in the history of philosophy, we have a solid explaination for what happened to these enlightened fordmakers, and indirectly, a formal proof of God.

[Karnika Kapoor · Rating 5 out of 5]

Touche!

[Thebookiemonster · Rating 2 out of 5]

I understand and mostly agree with the author's assertion that conclusions 1, 2 or 3 must be true. But I think the rest of the paper is wild.

Basically, he says future civilizations, if there are any, will have super powerful computers capable of simulating fake worlds. Naturally, those civilizations would then use those super powerful computers to simulate fake worlds. Uhh, I don't follow that. He also never addresses the purpose of running such simulations. I guess it would just be fun.

It's also possible the future civilization already exists and has created the entire universe we currently live in as a simulation. OK, that would be hard to disprove. The whole thing strikes me as a neat thought experiment rather than a theory.

To be fair, this is clearly not intended for the layman and it uses a lot of terms only those familiar with the literature would recognize.

[Logan · Rating 4 out of 5]

4 stars

When I first started reading this paper, it seemed just a bit too technical for me. Words like "doomsday theory" and "belief-states" made a bit nervous about my choice for today. However, once you get past the probability theory and the computational calculations, this philosophical points made are quite compelling and make you think deeply about the world around you.

The three points it aims to show, from the abstract, are that either:

1) we will not reach a post-human state;
2) we won't have any interest in running simulations; or,
3) we are almost certainly living in a simulation.

The paper presents in a rather understandable way, and now I understand just why so many luminaries of the day like to point to this paper for others to read.

[Barrett · Rating 4 out of 5]

https://www.simulation-argument.com/s...

[Noa MacDonald · Rating 5 out of 5]

I read this a good while ago, and I wonder if it’s a bit trite to put academic papers on Goodreads — though why books and not papers amounts to probably something superficial — but anyway, this popped up and frankly it’s one of the most compelling papers I’ve ever read, so to hell with it. It also pays to read people’s objections to this paper, and Bostrom’s replies: I think when all is said and done, on minimal or perhaps no patching, the argument for the trilemma stands.

Not to be confused, of course (though confusion abound) with the simulation hypothesis, which is just the affirmation of the third lemma, however one cashes that out. Nor to be confused with the (perhaps misleadingly named) simulation realism of Chalmers, which is an epistemologically-oriented (but not skeptical) take on the simulation hypothesis that consists in roughly, chaining “we cannot know we are not in a simulation” to the negation of “…and so, nothing is real.” On second thought, perhaps not misleadingly named?

The other thing I’ll say is that this paper, perhaps not by itself — simulation literature is very old — but certainly for me most cleanly makes the case for a naturalistic creator. Said a different way: the simulation argument posited here and its offshoots (like simulation realism) make the most convincing case against metaphysical atheism I’ve yet read (now of course it depends what one takes atheism to mean; here “belief in a creator” will do). And so I think Chalmers is right here: there is a brand of “simulation theology” that makes a plainly coherent pitch for a naturalistic creator who is mostly omnipotent and omniscient (jury’s out on benevolence), in the “next universe up,” as the phrasing goes; a creator who, well…created our universe. In fact, it’s been argued (by the philosopher Richard Brown) that all arguments for theism, save the ontological argument of Aquinas, are better arguments for simulation (theology), and that strikes me as roughly correct.

[· · Rating 2 out of 5]

(9 April, 2025)

From computational power to understanding Transhumanism, Bostrom starts a compelling discussion. A few more points to consider would be welcome, but he feels brevity is his friend here. A single mention of quantum computers is a mistake, as is only tangentially commenting on P-zombies (without naming them!). The religious implications of an omniscient simulator, or simulators, leaves me wondering about many 'What if' scenarios, all of which make me uncomfortable.

The assumption of posthumans constituting a monolithic entity is a strange one, a single 'individual' is all that is needed to undo the premise. If that separate posthuman is capable, and willing, to run said simulation, it doesn't matter what the 'others' think or do.

Fine as arguments go but not super stimulating. I already believe in the possibility we are all living in a simulation but few of my reasons for it are included here.

[Ahmed · Rating 5 out of 5]

So we are living in simulation. It doesn't seem so far-fetched.
Occam's razor shall apply once we reach posthuman stage and be able to run our first simulation, then the assumption that we are living in the matrix will get a closer to a 100% probability.
Nick also included some interesting hypothesis that can make the math even easier. For example, there is no need to simulate an entire history. just the current moment with fake memory for the past history.
I have another assumption not addressed by Nick: what if our technical limits are not the same as in our simulating lord universe. He might be in a universe with a much smaller Planck constant, a much higher speed limit, etc. Operations that sound ridiculous to us like 10^33 logic operations to simulate human history might be achieved on a mobile phone on their world.

[Astragen · Rating 1 out of 5]

If you're into stuff like this, you can read the full review.


Blatant Contradictions: "Are You Living In a Computer Simulation" by Nick Bostrom


Glad to see Bostrom is able to make a living out of total nonsense. If we’re in a simulation how did I end up playing this shitty part? Isn’t there a way of doing a challenge and upgrading to a playboy type character? According to Bostrom, every simulation has its own rules, i.e., you need to find out what they are, and then you can level up. In this simulation, apparently the rules are to accumulate large amounts of money and impoverish everyone else.

[J_BlueFlower · Rating 4 out of 5]

A classic by now. Has inspired a lot of fiction. I was a bit surprised about how many of the aspects are in this original paper. Like: “Simulating even a single posthuman civilization might be prohibitively expensive. If so, then we should expect our simulation to be terminated when we are about to become posthuman.) “

I am not a great fan of his calculations. The same could have been said with ordinary words. Feels a bit like the joke “lacking rigour? Rewrite in LaTeX”

[Brian Mikołajczyk · Rating 4 out of 5]

If technology persists, we will eventually have computers powerful enough to be able to simulate a civilization. Are we already in a simulation that our ancestors made? Bostrom uses probability and philosophic principles to answer this question. If humanity is to make a posthuman civilization then odds are we are indeed all in a simulation.
A very interesting read!

[dashiell h · Rating 5 out of 5]

At the moment, I’m a dogmatic nonbeliever in this supposed substrate independence proposition, at least at the level that Bostrom proposes is minimally necessary. Accordingly, we’re probably either in a simulation or never gonna simulate anything? Fun read; Reminds me of that one Veritasium Youtube video about the sleeping beauty paradox -> https://youtu.be/XeSu9fBJ2sI?

[Arturo · Rating 5 out of 5]

It is a very concise and well-explained article that might blow your mind about the possibilities, for some might be disturbing and for others, an amazing reference but either way it is a must-read for everyone no matter your interests, it is about our reality and we should be aware of it.

[Steven Gripp · Rating 5 out of 5]

Complementary materiel to simulation theory/Gnosticism. I’m not going to explain what he says, but what I gathered is his speciation on what simulation would be founded upon and where that existence should be recognized, if the possibility of a simulation occurred.

[Medhat · Rating 5 out of 5]

Complex argumentation of a computer simulation, similar to brain in a bat hypothesis much better, bostrom's computationalism theory seems to be enigmatic and philosophy bourgeoise and complex to digest

[Medhat ullah · Rating 4 out of 5]

for some reason a complex paper "subtrate independence", "indiffernce principle", "doomsday argument" start to popping up, well-versed and complex paper to read for a layman but "functionalism", "computationalism" :)

[Paula · Rating 5 out of 5]

Eu não acreditava até começar a ler, mas agora tudo fez sentido, chega a ser ridículo de tanto sentido que fez.

[Eric · Rating 4 out of 5]

My takeaways:
If we accept that 1) consciousness is substrate independent and 2) computational power will continue increasing, the argument that we are in a simulation is quite compelling.

[Vivek Vikram Singh · Rating 4 out of 5]

Brief and brilliant. Case 2 could have been argued more hence the deduction of 1 star.

[Rory Tregaskis · Rating 4 out of 5]

Did something strange to my brain I'm not sure I can undo.

[bia · Rating 5 out of 5]

the theogonical implications part was my favourite... maybe it's just simulations and simulators all the way down.