The „Matrix“ movies have made the idea that we might be living inside a computer simulation popular. The philosopher Nick Bostrom published in 2003 a theory (the „simulation hypothesis“) which states that it is well possible that we are living inside such a simulation. There are many ideas how such a simulation could be implemented technically. In the following we will explore some of them. Please do not take the following discussion too seriously. It should rather be a fun exercise in kitchen table philosophy.
- „Brain in a jar“.
In this variant reality is simulated only for one person. This is technically the easiest to implement. It would be sufficient to immerse the brain (which could also exist only as a computer simulation, i.e. without biological hardware) into some kind of hyperrealistic computer game. This is relatively easy, because only the parts of the world which are perceivable by this single person would have to be simulated. This world would include other living beings (e.g. your friends and your dog etc.), but only when they can be perceived. Also, because of the limited processing power of a single human brain, the simulation would not have to be very consistent. In „Matrix“ terminology: we are poor at noticing glitches in the „Matrix“. This could be exploited to simplify calculations. This is what is already used in 3D engines for computer games, where hundreds of (more or less) sophisticated mathematical tricks are used to fool the observer to accept the presented reality as real. Already in todays 3D games, we see - for instance - very realistic complex explosions of objects without the need to simulate them at the level of atoms and chemistry. It is safe to assume that such a simulation will be possible even for us humans quite soon (like in a decade or two). - „World simulation“
In this case the world (but not the entire universe!) as it is perceived by all humankind (and maybe the animals too) is simulated on a computer. Like in the first version, a simplified model of the world could be used to save computational resources. But as the world would have to allow consistent interactions of the whole population, the computational requirements are much higher than in the first case. Again we could have connected biological brains (i.e. the brains „run“ directly on the hardware of the host universe. This is the version envisioned in the „Matrix“ movies) or the brains are also simulated (like in the „Black Mirror“ episode „San Junipero“). In the latter case the computational requirements are even higher. - Universe simulation with limited computational resources.
It is also possible that the whole universe is simulated and not only our planet. The simulation would happen on the level of fundamental particles. The computational resources required to do this must be enormous. Therefore most people assume that in this case some tricks must be used too to keep the requirements at a tolerable level. In this case, it might be possible to detect imperfections in the simulation resulting from such tricks. We could actually prove that we are living in a simulation. This is the version which is usually discussed by experts and philosophers today. - Universe simulation with unlimited computational resources.
If we assume that the entity which is running the simulation (a „God“ or an extremely advanced, godlike civilization) has unlimited computational resources available, the simulation could be absolutely perfect. In this case it would be impossible to detect glitches and we could never know that we are living in a simulation. I will explain in the following, why I believe that this version could be possible and even more likely than the other variants.
How could it be possible that an entity has unlimited computational power available? I think the idea of limited resources is a very human one. It comes from our experience of growing up on this finite planet where we struggle all life competing for resources with others. Human life is limited mostly by the limited availability of matter. And what we humans can do with this matter is very limited as well. We know this since some early chemists tried to create gold by boiling down large amounts of urine. It did not really work (they discovered the element phosphor instead). So for some time we believed that matter is made from different kinds of atoms (from the greek word ἄτομος/atomos - the undividable) which cannot be altered. But later we learned that atoms have in fact a nucleus which can be split. Therefore a limitation which was believed to be fundamental suddenly disappeared and made a large number of new technical inventions possible. This means, it could well be that all the limitations we experience today vanish for a sufficiently advanced entity. That in a complete theory of physics only imagination is limiting what can be made to exist. In this case, the size and complexity of simulated universes would not matter anymore. It could be feasible and convenient to simulate a whole universe only to generate a few unique living beings.
One could argue that it could be technically challenging even for a highly advanced entity to run a simulation of a universe within another (kind of) universe. We know that simulations of old computer hardware (i.e. an Apple II) on modern computers are much slower than if we run the same algorithm directly on the modern hardware. But maybe the approach used resembles more something like containers (OS-level virtualization): there could be only a thin isolation layer between the host universe and the simulated universe and the simulated universe could run in some way directly on the hardware of the host universe. In a way we are exploring such technology already today on a very limited scale in the form of quantum computers.
Why is this version more likely than the other variants? 1 to 3 represent technical implementations which are used by entities with limited resources and therefore a lower level of development than an entity using method 4. If we accept that 4 is possible, it must be much more likely: it makes sense to assume that an entity which is capable of running a simulation of type 1 to 3 would progress quickly to a civilization capable of type 4 simulations. Why does this assumption make sense? We know from our history that technical knowledge grows at least exponentially (maybe there will be even a singularity). In this case, the time window for 1 to 3 simulations is very small compared to the very long time which comes after the transition to type 4 capability. Therefore, if all 4 types are possible, then the 4 type is by far the most likely.
From Bostrom‘s argument and the arguments in this article we can „conclude“ (with a large spoon of salt) that it is well possible (if not likely) that a godlike entity is „running“ our universe. Whether this entity is revealing itself to us or not, is a matter of personal belief.
Image: Shutterstock / thinkhubstudio
Follow me on X to get informed about new content on this blog.
