It might seem obvious that the world around us is real. Yet as computers have become ever more powerful, it seems increasingly possible that everything we experience could be nothing more than code running on the hardware of the future.
All we can rely on to explore reality is the information that we receive through our senses. But we know these can be fooled.
Every object we see as solid is in reality made up of atoms, and those atoms are mostly empty space that is anything but solid. Worse still, our senses regularly mislead us, as optical illusions demonstrate so brilliantly.
What, then, if everything we experience is trickery?
Fiction and philosophy have both delivered scenarios where the truth of reality bears no resemblance to our sensory experience of it. And with the development of computing, the concept of reality being a digital simulation has come to the fore.
The 1999 film The Matrix was not the first example by any means, but it provides a vivid example of an imagined world where every apparent experience is fed into the brains of humans held in pods.
The idea that our reality could be virtual was termed the ‘simulation hypothesis’ by English philosopher Prof Nick Bostrom, based at the University of Oxford, in the paper Are You Living in a Computer Simulation, published in 2003.
Bostrom imagines a future (based, he admits, as much on science fiction as the work of “serious technologists”) where vast amounts of computer power are available, and our descendants (whether for fun or research) run complex simulations, so detailed that billions of individuals within the simulation have a form of consciousness.
Bostrom primarily looked at such ancestor simulations, where our distant descendants run simulated versions of the kind of life lived in their past, though others go beyond this to consider simulations run by alien civilisations, or even beings living in another universe with different physical constants to those present in our simulation.
In The Matrix, there were two forms of existence: the virtual reality of a fake late 20th century that the central character Neo experienced for most of his life; and the real world, the scuzzy reality he discovers himself in after he takes the red pill.
Philosopher Prof David Chalmers of New York University argues that this isn’t the right way to view the simulation hypothesis. Because the circuitry and data in the computer the simulation runs on are real, he argues that simulated life is just as authentic as any other – just constructed from computer chips and circuits rather than neurons.
But most of us feel that physical existence provides a different level of reality from the virtual alternative.
Read more:
- 5 illusions that reveal how your brain warps reality
- What does the world look like outside of our brains?
- The 7 biggest myths about your reality, busted by science
The Truman Show limit
However, there is one problem with such simulations – energy. Italian astrophysicist Franco Vazza raised the energy issue in his 2025 paper Astrophysical constraints on the simulation hypothesis for this Universe: why it is (nearly) impossible that we live in a simulation.
Here, Vazza examined three scenarios: simulating the entire visible Universe, just simulating Earth, or simulating a “low-resolution simulation of Earth compatible with high-energy neutrino observations,” – essentially the minimum detail required to match existing experiments.
He concludes that there is not enough energy in the observable Universe to simulate a full universe to absolute detail.
To simulate the Earth, he suggests that it would be necessary to convert the entire stellar mass of a globular cluster into energy. And while in principle the low-resolution version is attainable, it might easily take millions of years to simulate a single second of Earth time. That’s a lot of energy for a full simulation.
Of course, there’s no reason why a computer simulation should have billions of humans fully replicated, or include stars that are more than points of light in the sky. It doesn’t have to be that complex.
We can learn something about the simplest possible simulation from the 1998 movie, The Truman Show.
Central character Truman Burbank, played by Jim Carrey, leads an apparently normal life, but in fact lives on a vast sound stage, with actors filling in for everyone he encounters. Truman’s reality is entirely fake – but only he believes the simulation.
So, it could be the Universe is a simulation, focused solely on the only person that you can be sure is a truly conscious being – you. There would be no need to simulate the Universe, or even the Earth. Your direct experience is extremely limited – and that is all that needs to be simulated.
Vazza admits this possibility but suggests that even this approach would collapse as soon as scientific experiments were undertaken to look at the Universe in more detail.
For example, if they were to look at something through a microscope, the simulation would have to include that new level of detail. If they looked at even smaller scales, the simulation would have to become ever more detailed, until it exceeded its processing limits.
He commented, “Although one can be not interested in physical details, still we are constantly relying on processes that happen on incredibly smaller scales (from atomic reactions to electromagnetic wave transmissions to light propagation, for instance).
“Even if we pretend we could ignore them, they surely must happen at a fundamentally small scale.”
Arguably, though, this would only be an issue if the single individual simulated were a scientist, interested in looking at the smaller details of the Universe around them.
Most people who could be the possible central subject of a simulation are not scientists. Their only experience of these measurements of the smallest scales of our Universe will be indirect, coming from people like Vazza.
Vazza admitted, “I did not consider a scenario without scientists – being one of them and being surrounded by them, it’s hard to remove them from the picture!”
A Truman Show simulation remains a real possibility – but if that is the case, I wonder if I should even be writing this article. As I know I’m conscious, there would be no one else in my simulation to read it.
Statistical chicanery
Let’s put aside the energy problems for now, assuming that some future civilisation has found a way to run a simulation sophisticated enough to simulate the reality of several billion people. How likely is it that we are actually living in a simulation?
In his original paper, Nick Bostrom put forward two reasons why an advanced civilisation might not be running a simulation:
- The fraction of human‐level civilisations that reach a posthuman stage is very close to zero, and so few reach the required technological level needed to run a simulation.
- The fraction of posthuman civilisations that are interested in running ancestor‐simulations is very close to zero, meaning no one wants to run a simulation.
He concluded that if neither holds, then we are likely to be in a simulation, because if there were many civilisations with the will and ability to run simulations, the number of simulated universes would outnumber the physically real universes.
There are two potential problems with this argument. One is that there are other reasons simulations might not be built than Bostrom’s two possibilities – there is a huge assumption, for example, that such a huge simulation is technically possible.
As we have seen, this may not be the case. The other issue is that the statistical argument employed here resembles the inverse gambler’s fallacy.
The original gambler’s fallacy is a natural feeling that probability should influence events. For example, if we toss a fair coin and it comes up ten heads in a row, we feel a tails should be more likely to even things out. But the coin has no memory. A heads is just as likely on the next flip.
For the inverse fallacy, imagine you walk into a casino. The first person you come across is winning time after time. A cosmologist friend comments to you, “The casino must be very busy.” Asked why, she says that because winning so often is very unlikely, there must be lots of other people playing.
This too is a fallacy because you can’t deduce anything about the population as a whole from one individual. Similarly, we can’t deduce that we are in a simulation simply because there are likely to be many more simulations than real universes.
Testing for simulated reality
If statistics can’t help, is there a way to distinguish between being in a simulated universe and a physical reality? There are certainly potential ‘glitches in the Matrix’ where our understanding of the Universe doesn’t fully fit together.
There have even been some suggestions that the finite speed of light could be a limitation of a simulation’s processing speed. Similarly, the fact that our Universe is made up of a collection of elementary particles could be a sign that our reality is ‘pixelated’. Or they could be just how a real universe works – we have no way of telling.
Short of waiting for the laws of physics to briefly fail, we need to be able to spot evidence that our existence is simulated.
I asked Nick Bostrom for suggestions. He replied, “There are clearly possible observations that would show that we are in a simulation. For example, the simulators could make a ‘window’ pop up in front of you with the text ‘YOU ARE LIVING IN A COMPUTER SIMULATION. CLICK HERE FOR MORE INFORMATION.’ Or they could uplift you into their level of reality.”
However, these are not tests that we can instigate – and there is no evidence that such interventions are likely.
As an alternative, Bostrom offered the possibility of obtaining indirect evidence: “The simulation hypothesis is empirically testable in the sense that there are possible observations we might make that would either increase or decrease the probability that it is true.”
Such observations would be about the chances of our civilisation surviving and wanting to create simulations – for example, if we identified hazards that would make it near-impossible for our civilisation to survive long enough for such technology to be developed – but it’s hard to see how this would be put into practice.
“I doubt that there is any simple experiment – like mixing some chemicals in a bottle and checking whether they turn red or blue – that would do that,” says Bostrom.
Prof Melvin Vopson of the University of Portsmouth’s physics department disagrees, suggesting that you can somehow detect the disappearance of the information contained in particles when they annihilate, which could give away that we are in a simulation. But this is not a widely accepted test.
Bostrom concluded, “Furthermore, if there were some such experiment, I’m not sure it would be a good idea to do it.”
After all, a simulation may no longer be worth running if its occupants know they aren’t real.
Where do we go from here?
Outside of energy limitations, there has been at least one attempt to use mathematics to disprove our potential simulated nature.
Although not the central topic of Consequences of Undecidability in Physics on the Theory of Everything by Mir Faizal, Lawrence M Krauss and others, the paper uses Gödel’s incompleteness theorems – which prove that there will always be unprovable statements in any mathematical system – to suggest that the Universe cannot be a simulation.
The authors argue that because there are aspects of the Universe that we can observe but that can’t be settled by an algorithm, we can’t be in a simulation. For example, it may be possible to see the behaviour of a black hole that isn’t describable mathematically.
However, this argument is purely theoretical: we don’t know if any existing scientific observations are not computable. Such arguments feel a little like medieval attempts to logically prove or disprove the existence of God.
To be a scientific hypothesis, there needs to be a mechanism to gather evidence for that hypothesis, even if it is, as yet, outside our technological grasp. It seems likely that, by this measure, the simulation hypothesis is not science.
You may never have a scientific method to determine whether your reality is anything more than processes within a computer program. But it won’t stop it being fun to ask.
Read more:
