The Question That Will Not Go Away
In 2003, Oxford philosopher Nick Bostrom published a paper that sent an uncomfortable idea rippling through academic circles: at least one of the following must be true. Either almost all civilisations go extinct before reaching the technological maturity to run detailed simulations of their own history. Or advanced civilisations have no interest in running such simulations. Or we are almost certainly living inside a computer simulation right now.
That third option has refused to stay quiet. In the two decades since Bostrom's paper, physicists, computer scientists, and cosmologists have found a growing and unsettling body of evidence suggesting that the universe behaves, in some surprisingly specific ways, like a computational system. None of it is conclusive. All of it is worth taking seriously.
The Universe Has a Pixel Size
One of the most striking features of physical reality is that it appears to be fundamentally discrete rather than continuous. At the smallest scale, space and time are not infinitely divisible. The Planck length — approximately 1.616 × 10⁻³⁵ metres — is the smallest meaningful unit of distance in physics. Below it, the concepts of space and distance break down entirely.
If you were designing a simulation and needed to conserve computational resources, making space and time granular rather than infinite would be exactly the right engineering choice. The fact that our universe has a natural resolution limit — a pixel size, essentially — is at minimum a striking coincidence with what we would expect from a simulated system.
Physicist James Gates at Brown University discovered something even more striking: error-correcting codes embedded in the equations of supersymmetric string theory — the same class of codes used in browsers and operating systems to detect and fix transmission errors. Their presence in the fundamental equations of physics is either a profound mathematical coincidence, or something more.
Quantum Mechanics: Reality Only Exists When Observed
The quantum mechanical behaviour of subatomic particles is one of the most well-confirmed and strangest results in all of science. Particles do not have definite positions or properties until they are measured — they exist in superpositions of states, collapsing to a definite value only upon observation. This is not a limitation of our instruments. It is a fundamental feature of nature, confirmed by decades of rigorous experiments.
From a simulation perspective, this behaviour makes perfect computational sense. A simulation only needs to render what is being observed. Keeping unmeasured particles in an indeterminate superposition until the moment of observation is precisely the strategy a resource-efficient simulation would employ — why spend processing power rendering detail that nobody is looking at?
Physicist Max Tegmark of MIT has argued that the mathematical structure of reality is so deep and pervasive that the universe may literally be mathematics — which would make it, in a fundamental sense, intrinsically computational.
The Speed of Light as a Processing Speed Limit
Every simulation has a maximum processing speed — a clock rate that limits how fast information can be transmitted across the system. In our universe, the speed of light plays exactly this role. Nothing — no information, no physical effect — can travel faster than approximately 299,792 kilometres per second.
Why should such a limit exist? From a purely physical standpoint, there is no obvious reason why space could not permit faster-than-light effects. But from a computational standpoint, a universal speed limit is precisely what you would expect: it is the maximum rate at which the simulation can propagate updates through its coordinate grid.
Cosmologist Paul Davies of Arizona State University has pointed out that the fine-tuning of physical constants — the precise values of gravity, electromagnetism, and dozens of other parameters — is so improbably perfect for the emergence of complex life that either there are vastly many universes with varying constants, or the constants were deliberately chosen. A simulation designer selecting parameters to produce interesting, evolving complexity would set them exactly this way.
The Fermi Paradox Gets a New Answer
The Fermi Paradox — the eerie cosmic silence despite the statistical near-certainty that intelligent life should exist elsewhere — has long baffled scientists. The simulation hypothesis offers a darkly elegant explanation: perhaps the simulation does not render what is not necessary for its primary purpose.
If the simulation is running to observe the development of one particular civilisation, a cosmos full of richly detailed alien worlds would be a colossal waste of processing resources. Better to leave the rest of the universe as a sparse backdrop — galaxies that look detailed when examined closely but are not actively simulated in full. The emptiness of the cosmos, on this reading, is not a mystery but a feature of efficient computational design.
The Case Against
Serious objections exist, and they deserve equal weight. Physicist Sabine Hossenfelder has argued forcefully that the simulation hypothesis is not science in any meaningful sense — it makes no testable predictions and cannot be falsified. A hypothesis that can accommodate any possible observation without ruling anything out is doing philosophy, not physics.
The computational resources required to simulate a universe with the fidelity we observe would be, by any estimate, incomprehensibly vast. And the concept of a "computer" running the simulation raises an immediate regress problem: what is that computer made of? Does it exist in yet another simulation?
There is also the point that our intuitions about computation are drawn entirely from machines we have built — crude and limited relative to whatever hypothetical system could run a cosmos. Analogising the universe to a desktop processor may be a category error of the most fundamental kind.
Does the Answer Matter?
Even if we could never know definitively, the question carries real philosophical weight. If we do live in a simulation, the traditional concept of physical reality as the ultimate ground of existence dissolves entirely. Questions about consciousness, free will, and meaning take on new dimensions that science and philosophy are only beginning to map.
Elon Musk has claimed there is a one-in-billions chance we are not in a simulation. Neil deGrasse Tyson puts the odds at better than 50-50. Most physicists are more cautious — but the fact that serious scientists engage the question at all marks a genuine shift in how we think about reality.
The simulation hypothesis will not be proven or disproven anytime soon. But it has done something genuinely valuable: it has forced a new generation of thinkers to ask, with fresh eyes, the oldest question of all. What, exactly, is reality made of?
For deeper reading, explore Bostrom's original argument, the Foundational Questions Institute, and the ongoing debate at Scientific American.