Algorithmic Idealism I: Reconceptualizing Reality Through Information and Experience

Imagine standing at the edge of a vast, quiet forest, trying to understand how the trees grow. For centuries, scientists have looked at the world from the outside, mapping the roots, the soil, and the sunlight to explain the forest's existence. This is the traditional view of physics: the universe is a stage, and we are the actors moving across it. But what if the stage itself is not the most fundamental thing? What if the only thing that truly exists is the actor's experience of the moment? This is the question that drives a new line of thinking called Algorithmic Idealism. It does not ask what the world is made of, but rather what an observer should expect to experience next. It suggests that the laws of nature might not be rules written on a cosmic blackboard, but rather a method for predicting the next step in a personal story. This approach tries to solve deep puzzles that arise when we imagine strange situations, such as a person being copied perfectly or a universe that creates random, fleeting minds out of chaos. By focusing on the immediate "self" rather than an external world, researchers are testing whether the reality we see is something that emerges from our own stream of consciousness, much like a pattern that appears when you look at a complex design from a distance.

The paper you are reading is a careful, updated assessment of this idea, written by Krzysztof Sienicki in August 2026. It serves as a reality check for a theory that was previously described with more confidence than the evidence allowed. The author revisits the work of Markus P. Müller, a physicist who proposed that the universe is fundamentally a series of "self states"—abstract snapshots of what an observer is experiencing right now. The core idea is that instead of assuming an external world exists and calculating how it changes, we should start with the observer's current state and use a universal method of prediction to guess what comes next. This method is based on the idea that the simplest explanations for patterns are usually the most likely to be true. If you have seen a sequence of events, the most probable next event is the one that fits the simplest pattern. The paper argues that if you apply this rule consistently, a stable, external world begins to appear as a side effect, even though it was never assumed to be there in the first place.

However, the 2026 revision makes it clear that this is not a finished solution to all the mysteries of the universe. The author distinguishes between what has been mathematically proven in a specific, simplified model and what is merely a hopeful guess about the real world. In the most successful version of this theory, known as the "bit model," the observer's state is treated as a string of zeros and ones. In this simplified world, researchers have rigorously proved that if you keep adding new bits of information, the sequence will eventually look exactly like the behavior of a real, physical world governed by standard laws of probability. This is a significant achievement because it shows, in a mathematical sense, how an "outside world" can emerge from a purely internal perspective. But this proof only works in a world where information is never lost. In the real world, we forget things, our memories fade, and data gets erased. The paper explicitly states that the current theory cannot yet handle this "erasure" problem. Until a version of the theory can explain how a world emerges even when information is deleted, the claim that it explains our actual universe remains incomplete.

The paper also corrects several misconceptions that had grown around the theory. One major point is that this framework is not a new way to interpret quantum mechanics, the physics of the very small. While quantum theory inspired the idea, the author is careful to say that Algorithmic Idealism is a separate project. It does not automatically explain why particles behave the way they do, nor does it prove that the famous "Born rule"—the formula that predicts the odds of quantum events—is the same as the prediction rule used in this theory. Instead, the theory hopes that in the right conditions, its predictions will match what quantum mechanics already tells us. If they do not match, the theory would need to be adjusted. Similarly, the paper pushes back against the idea that this theory has solved the "Boltzmann brain" problem. This is a paradox where random fluctuations in the universe could theoretically create a brain with false memories more often than they create a whole person living a normal life. While the theory suggests that a person should not simply assume they are one of these random brains just because there are many of them, the paper admits this is a specific prediction of the model, not a proven fact that eliminates the problem for all of cosmology.

Another area where the theory offers a fresh perspective is in the realm of personal identity, such as when we imagine teleporting a person or copying their mind into a computer. The theory treats the "self" as a pattern of information rather than a specific physical body. If two different physical machines hold the exact same pattern of information, the theory says the future experience of the "self" is the same for both. This challenges the idea that there is only one "real" person and the rest are just copies. However, the paper is careful not to claim this solves the philosophical question of whether the copy is truly the same person. It simply provides a rule for calculating what the observer should expect to happen next, without settling the deeper metaphysical debate about who they are. The author also notes that the theory does not automatically tell us how to treat these digital or copied beings ethically. Knowing what is likely to happen does not tell us whether it is right or wrong to do it.

The most significant development highlighted in the paper is a new kind of scientific debate called an "adversarial collaboration." In 2026, the theory's creator worked directly with a critic who holds the opposite view: that the physical world is fundamental and real. They agreed to test their ideas against each other to see where the theory holds up and where it breaks. This collaboration has not settled the argument, but it has clarified exactly what needs to be proven. The central challenge now is to create a version of the theory that works when information is lost, to show that it can reproduce the known laws of physics without just copying them, and to prove that its strange predictions about teleportation and simulations are robust. The paper concludes that Algorithmic Idealism is a promising and mathematically sophisticated research program, but it is not yet a completed theory of everything. It is a work in progress that has successfully shown how a world can emerge from a single perspective in a simple model, but it still has a long way to go before it can explain the complex, messy reality we live in. The value of the work lies not in having all the answers, but in asking the right questions about how we experience the universe and how we can test those experiences against the laws of nature.