Sometimes when certain ideas are trending the scientific press, it's well deserved, a lot of the time it's not. I imagine by now most people have probably heard of the Simulation Argument, it comes from a 2003 paper by the philosopher Nick Bostrom. Essentially it's a statistical argument to the affect, that we're very likely living in a simulated universe rather than a real one. The argument goes something like this:
1. It's possible to simulate a universe like ours, with the kinds of experience we have in this world.
2. Perhaps far in the future, the technology to do so will exist and become sufficiently wide spread.
3. Future beings will be interested in running ancestor simulations of their past.
4. If 1-3 are true, then in the entire history of the universe simulated worlds will vastly out number real ones.
We should assume we’re typical observers and so therefore we are likely living in a simulated world rather than a real one. A lot of people like the idea that the universe is a simulation, they may think it speaks to their mental flexibility but actually it portrays a fairly weak grasp of physics. The argument was covered on Sean Carroll's blog and he has a pretty good objection to it.
One way to understand this picture of the world is in terms of "digital physics" that all of reality, the laws of nature, spacetime, matter, energy are at bottom made up of grid nodes of bits, and encoded on an algorithm; looking at it top-down, gravity is the curvature of spacetime, so a quantum theory of gravity entails that spacetime is made up of discretely quantized parts.
That's a very naive approach to quantum gravity, and today we know a lot better. If we try to build up spacetime from 'atomic-like chunks' then our theory is not going to be Lorentz invariant–that's fancy sounding science jargon for "in conflict with known physics" namely special relativity–if with respect to one reference frame I cannot talk about time separations smaller than a certain duration, then I can have a second frame of reference moving very fast with respect to first where those separations become enormously large. Where violations in the principle of relativity will be observed.
Second, if spacetime is built up from Planck sized units of volume, I should be able to do a very rudimentary calculation to infer the total number of bits of information I need to specify for a complete picture of that given region of spacetime. It should be:
The problem here is that, that's false. We have a better way to calculate the information encoded in a region of space, using the Holographic Principle. The degrees of freedom are limited by the area of the boundary of a given region of spacetime, so S is proportional to the area, not the volume of that region.
Certain features of our world–spacetime, potentially consciousness and in my view, God–cannot be simulated on a computer, that already makes the Simulation Hypothesis impossible to accept.
I wish science popularizes (or the science media) would stop promoting the simulation hypothesis as though it were serious science, it's not even close to being half of a scientific idea. Digital physics hasn't derived General Relativity or the Standard Model of Particle physics, and there are seriously questionable philosophical assumptions fed into the argument as well.
It neglects for example that the physical substrate for consciousness is a biological organ, not an algorithmic code. Bostrom assumes "substrate independence" but the fact that I can simulate something looking very similar to rain on a computer doesn't mean that the computer program is itself wet, inside. Even if you could simulate a conscious being neuron-by-neuron, it would be at least as complex as the brain itself. Each person can't be "copied and pasted" but is an individual, which makes it incredulous at best, to imagine that simulated people will hugely out number real people.
The simulation argument isn't something that keeps me up at night, most people who read about these ideas in the scientific press are probably going to conclude that these "science" authors are full of shit and I don't blame them.