Not a movie plot — a 2003 philosophy paper
Pop culture usually traces this idea to The Matrix, but the specific, rigorous version of the argument predates and outlasts the movie. In 2003, philosopher Nick Bostrom (University of Oxford) published a paper that doesn't claim we're in a simulation at all — it's a logical argument about probability, built on an assumption that's easy to miss in the retelling: that a sufficiently advanced ("posthuman") civilization could eventually run simulations detailed enough to contain conscious beings who don't know they're simulated.
What the paper really argues, in his own words
Bostrom's paper, "Are You Living in a Computer Simulation?" (Philosophical Quarterly, 53(211), 243–255), lays out three possibilities and argues that at least one of them must be true:
- The human species is very likely to go extinct before reaching a "posthuman" stage capable of running such simulations; or
- Any posthuman civilization is extremely unlikely to run a significant number of simulations of their evolutionary history; or
- We are almost certainly living in a computer simulation.
What gets lost in almost every pop-science retelling is Bostrom's own position on which of the three is true. He doesn't pick the third one:
“I do not accept this: I think we currently lack grounds for eliminating either of the three disjuncts.” — Nick Bostrom, simulation-argument.com
In other words, the actual paper is an argument that you can't currently rule any of the three out — not an argument that we probably live in a simulation. That distinction is the single most important correction this page exists to make.
"It from bit" and "the universe is a computer" aren't Bostrom's argument
A separate line of thinking gets conflated with Bostrom's argument constantly, but it's making a genuinely different claim. In 1989, physicist John Wheeler coined the phrase "it from bit", proposing that every physical thing — every particle, every field, even spacetime itself — ultimately derives from binary, yes-or-no information at its foundation. This is a claim about the deep structure of physical law, not a claim that the universe is being simulated by someone else's computer running in some outer, "base" reality.
Seth Lloyd (MIT), in Programming the Universe (2006), goes further, describing the universe itself as a quantum computer that computes its own evolution. That's still a different claim from Bostrom's: Lloyd's universe computes itself; Bostrom's hypothetical scenario is a universe being run by an external, more advanced civilization's computer. Keeping these two ideas separate matters — they get merged into one vague "we're in a simulation" claim constantly, but they're not the same argument and don't share the same evidence.
The most commonly cited piece of "evidence" — examined
The single most repeated claim connecting physics to simulation theory is some version of: "the double-slit experiment shows reality only renders in detail when it's being observed, like a video game only loading what's on screen." This page's companion piece, The Double-Slit Experiment, Explained, covers the real mechanism in detail — the short version is that the actual physics (decoherence through entanglement with any measuring apparatus, not consciousness) doesn't require or particularly support a rendering-on-demand interpretation.
Worth being precise about: this specific "video-game rendering" framing is a real, widely circulated claim — but after direct research for this page, no credentialed physicist appears to make this exact claim on the record. It functions as a pop-science/social-media version of a much more careful, abstract argument (Wheeler's "it from bit," above), stretched past what it can actually support. Even Elon Musk's well-known public comments on simulation probability lean on this kind of loose double-slit framing without engaging with the real decoherence physics — a useful example of how casually this specific claim circulates, not a scientific source for it.
A real mathematical discovery, already covered on this site
The second most commonly cited piece of "evidence" is physicist James Gates' 2008 discovery that certain supersymmetry equations contain structures mathematically identical to error-correcting codes — the same general kind of math used to check and correct digital data. This is a real, published, peer-reviewed finding. It's also a finding whose own discoverer explicitly rejects the simulation-theory reading of it, and whose foundation (supersymmetry itself) remains experimentally unconfirmed. Full details, sources, and Gates' own quoted caveat are on this site's Error-Correcting Codes in Physics page.
“Pseudoscience” — a specific, citable argument against
Physicist Sabine Hossenfelder has been one of the most direct public critics of simulation theory, calling it pseudoscience in a widely read 2021 post. Her argument isn't a vague dismissal — it's specific:
“Nobody presently knows how to reproduce General Relativity and the Standard Model of particle physics from a computer algorithm.” — Sabine Hossenfelder, Backreaction, 2021
She also directly addresses the "it only renders what's observed" escape hatch that simulation-theory proponents sometimes reach for, drawing a comparison to climate modeling: you can't simply discard short-distance detail in a simulation without that omission eventually corrupting the long-range results in ways that should, in principle, be detectable. Her conclusion is that proponents are making large, unproven assumptions about mechanism — "requiring faith rather than logic," in her words — not describing a testable scientific hypothesis.
As recent as 2025, and still not settled either way
This isn't old news sitting unchallenged. In June 2025, physicists Mir Faizal and Lawrence M. Krauss, with collaborators, published a paper arguing that Gödel's incompleteness theorem proves reality requires a kind of "non-algorithmic understanding" that no computation could ever reproduce — concluding, in press coverage, that the universe "cannot be, and could never be, a simulation." It ran in the Journal of Holography Applications in Physics, a minor, specialized venue rather than a flagship physics journal, which is worth knowing when weighing how settled the claim actually is.
Within months, a direct rebuttal appeared on arXiv (Evan Redden, "Provability vs. Execution: A Comment on 'Consequences of Undecidability in Physics on the Theory of Everything'"), arguing the original paper conflates two different things — what can be mathematically proven about a system versus what can actually be computed or executed — using Conway's Game of Life as a working counterexample of a system that's formally undecidable in places yet still runs and produces definite outcomes.
Separately, physicist David Wolpert (Santa Fe Institute) published a 2024 formal framework using the Church-Turing thesis and tools from computer science theory to even rigorously define what a "simulation" claim between universes would mean — concluding it's formally possible in some configurations, while other configurations run into real mathematical impossibility results. And astrophysicist Franco Vazza argued in 2025 that simulating our universe would require more energy and computation than is physically available under our own physical laws — an argument he himself notes only rules out simulation by a civilization bound by our physics; if the "outer" simulating universe runs on different physical laws entirely, he states plainly that the question becomes untestable, even in principle.
That a claimed mathematical disproof (Faizal, Krauss et al.) was directly disputed within months, by researchers working from a different formal angle (Wolpert) arriving at a genuinely mixed result, and by a third physicist (Vazza) concluding the question may be permanently untestable under certain conditions, is itself the most honest fact available here: this is not a settled question, in either direction, as of this writing.
What this is, and isn't
The simulation hypothesis, taken seriously, is a real philosophical and physics argument with real researchers on multiple sides of it — not a settled scientific finding, and not something either proven or disproven by any single piece of evidence commonly cited for it. Bostrom's own paper is far more modest than its pop-culture reputation suggests. The double-slit experiment's "observer effect" has a well-understood, non-mystical explanation that doesn't require a simulation to make sense of it. The error-correcting-codes finding is real but explicitly disclaimed by its own discoverer. And the most recent attempt at a mathematical disproof is, itself, actively disputed in the literature. The honest position, as of today, is that this remains a genuinely open question — which is a less exciting answer than either "yes" or "no," but it's the accurate one.
Other edge-of-knowledge topics, told the same way
This page is part of Resonant Inquiry, an ongoing series on this site covering genuinely strange, real research — sourced, with honest caveats, and no claims of knowing more than what's actually been shown.