In most of the world, the simulation hypothesis is a dorm-room conversation that physics politely declines to enter, because it is untestable by definition. Then Melvin Vopson, a working physicist at the University of Portsmouth, published a peer-reviewed paper proposing a falsifiable law of information that behaves, by his own argument, exactly like a universe optimizing its own code. That is the inversion worth taking seriously: not a philosopher speculating, but a physicist offering an equation and, crucially, an experiment.

Here is what actually happened, in the literature, with a citation you can check. The paper is titled 'The second law of infodynamics and its implications for the simulated universe hypothesis.' It appeared in AIP Advances, volume 13, issue 10, article 105308, in October 2023, with the digital object identifier 10.1063/5.0173278. AIP Advances is a peer-reviewed journal published by the American Institute of Physics. This is not a blog post or a preprint nobody vetted. It cleared review at a real physics journal, which is exactly what makes it a different kind of claim than the usual simulation talk.

The core idea runs directly against intuition. The second law of thermodynamics says entropy, disorder, tends to increase in an isolated system over time. Vopson, building on earlier work with collaborator Serban Lepadatu, says he expected the entropy of information-bearing states to do the same, to climb. Instead, across the systems he modeled, the 'information entropy' stays constant or falls. He named this the second law of infodynamics: in information systems, entropy does not increase, it remains stable or decreases over time, the mirror image of the thermodynamic law. He reports the pattern showing up across digital storage, genetic mutation data, atomic physics, and cosmological structure.

Why does that point at a simulation? Because a falling information entropy is the behavioral signature of compression and optimization. Any engineer building a vast simulation would face an impossible storage and compute bill and would do exactly what a video codec or a file system does: deduplicate, compress, prune redundant states, push the system toward minimal description length. A universe that spontaneously minimizes its information content over time looks, from the inside, like one running an optimization routine to keep itself cheap to run. Vopson explicitly frames it this way: built-in data optimization and symmetry are precisely what you would expect if reality were computed rather than merely physical.

Now the evidence, and the part that separates this from philosophy. Vopson did not stop at metaphor. He has proposed an actual laboratory test of a related conjecture, the mass-energy-information equivalence principle, which predicts that a bit of information has a tiny physical mass. The proposed experiment uses electron-positron annihilation: if information stored in elementary particles carries mass, the energy released when matter and antimatter annihilate should differ measurably from the standard prediction by an amount tied to the particles' information content. That is a falsifiable, do-it-in-a-lab claim. A simulation hypothesis you can disprove with a particle detector is a genuinely new object in this conversation.

The skeptical reading is necessary and substantial. Many physicists are unconvinced that 'information entropy' as Vopson defines it is the same quantity as physical entropy, or that its decrease implies anything cosmological rather than reflecting how he set up his information systems. Critics note that one peer-reviewed paper is the start of a conversation, not a verdict, and that 'looks like optimization' is suggestive, not proof. The proposed annihilation experiment has not, to date, returned a confirming result, and extraordinary claims will require independent replication before anyone should update their picture of reality.

All of that is fair, and Inverted World holds to it. But notice what has quietly changed. The simulation question used to be unfalsifiable by construction, which is the real reason serious science ignored it. Vopson's move is to drag it out of metaphysics and attach it to a measurable prediction about the mass of a bit. He may be wrong. The point is that he made it the kind of claim that can be wrong. The unresolved question is no longer the philosophical one. It is empirical and almost shockingly concrete: when matter meets antimatter, does the energy that comes out know how much information was inside, and if a careful experiment ever says yes, what exactly have we been living in?