Quantum Entanglement

Measurement Problem

Here is the puzzle that has consumed physicists for a century. Wave function evolves smoothly and predictably according to Schrödinger’s equation. Given a wave function now, you can calculate what it will be at any future moment with perfect precision. Deterministic. Elegant. But then you measure something and the wave function appears to collapse. From a spread of possibilities to a single outcome. Instantly. This collapse is not described by Schrödinger’s equation. It is bolted on as a separate rule. Why should measurement be different from any other physical process?

i Measurement: from many possibilities to one definite outcome ×

Schrödinger illustrated the absurdity with a famous thought experiment. Place a cat in a sealed box with a vial of poison triggered by a quantum event, say radioactive decay of a single atom. Quantum mechanics says atom is in superposition of decayed and not-decayed until measured. If you apply same rules to entire system, cat is in superposition of alive and dead. Obviously cats are not both alive and dead. So where does quantum superposition end and definite classical reality begin? This boundary, if it exists, has never been found. Every experiment designed to find a size limit for superposition has instead confirmed that larger and larger systems can exist in quantum superposition. The current record is a molecule of about two thousand atoms, sent through a grating in 2019 and showing interference like any other wave. The boundary keeps retreating.

What "measurement" actually means in physics remains genuinely unresolved. Is it consciousness? Interaction with a large system? Information becoming irreversibly recorded? Different interpretations of quantum mechanics give radically different answers, and all of them make identical experimental predictions. Physics cannot currently tell them apart. That is not a failure of physics. It is a clue that something very deep about reality remains to be understood.

Entanglement

Two particles can be prepared so that their wave function cannot be separated into independent pieces. They share a single quantum state. Measure one and you instantly know something about the other, regardless of distance. This is entanglement, and it troubled Einstein so deeply that he called it spooky action at a distance and spent years arguing it proved quantum mechanics was incomplete.

The thread in that animation is the way the idea sits in most people’s heads: a line running between the two particles, tugged at one end and felt at the other. The animation’s own caption already refuses it – it says one shared state, not two things joined – and the thread is worth looking at precisely because it is wrong. The rest of this page is an argument against it. Here is the first thing wrong with it. A thread joins two things that each have properties of their own. An entangled pair is the opposite case. The pair has a completely definite state – there is nothing further to know about it. Each particle taken by itself has no state at all: ask what its polarization is and the honest answer is that it does not have one, and every measurement you make on it comes out a coin flip. All of the information is in the pair and none of it is in the halves. That is not a limit on what you can find out. It is what being entangled means, and it is the actual measure: the less a particle knows about itself, the more strongly it is entangled with its partner.

Consider a pair of entangled photons created together with correlated polarizations. Before measurement, neither photon has a definite polarization. Both exist in superposition. Measure one and find it horizontally polarized. Instantly, no matter how far away, other photon’s polarization is determined to be vertical. How? The mathematics says they share a single wave function that cannot be factored into independent parts. But wave function is a mathematical abstraction. It does not explain what physical mechanism, if any, connects two particles across arbitrary distance. What measurably differs between entangled and non-entangled photon pairs is the pattern of correlations in their measurement outcomes. Entangled pairs show correlations too strong to be explained by any shared classical information they could have carried from birth. Where this correlation lives, how nature enforces it across space, is precisely what remains unexplained. Every experiment confirms entanglement works at every distance tested, from laboratory benches to satellite links spanning over a thousand kilometers. No distance limit has been found. But saying "they were never separate" describes the mathematics, not necessarily the physical reality underneath.

There is a second thing wrong with the thread, and it is the one that settles the matter. Threads can be strung freely: tie a particle to a second, then to a third and a fourth, and nothing stops you. Entanglement does not work that way. Coffman, Kundu, and Wootters proved in 2000 that it obeys a strict trade-off, and the extreme case is stark. If two particles are maximally entangled with each other, neither can be entangled with anything else in universe, at all. Entangle a particle a little with one partner and a little with another, and the two shares must add up to no more than it has to give. Physicists call this monogamy, and it is the behavior of a quantity that gets divided rather than a link that gets tied. It is also what makes quantum key distribution possible: if two parties can confirm they are maximally entangled with each other, they have thereby confirmed that no third party is entangled with either of them, because there is nothing left over to be entangled with.

Einstein, Podolsky, and Rosen argued in 1935 that this instant correlation meant quantum mechanics was missing hidden variables, secret internal properties that predetermined outcomes before measurement. Photons would carry hidden instructions from birth: "if measured this way, give this result." Correlation would then be like pairs of gloves shipped to different cities. Finding a left glove tells you the other box has a right glove, nothing spooky about it. Quantum mechanics, they argued, was merely an incomplete description of a deeper, deterministic reality.

Bell’s Theorem

For 30 years, EPR argument seemed like philosophy, a debate about interpretation with no experimental consequence. Then in 1964, physicist John Bell proved something extraordinary. He showed that any hidden-variable theory obeying locality (meaning no faster-than-light influence) must produce correlations that satisfy a specific mathematical inequality. Quantum mechanics predicts violations of that inequality. The predictions are different. You can test which is right.

i The angles are chosen at random at the last moment, and that is what makes the test work. ×

Starting with John Clauser’s experiment in 1972, sharpened by Alain Aspect in 1982, and culminating in loophole-free tests in 2015, results are unambiguous. Bell’s inequality is violated. The 2022 Nobel Prize in Physics went to Clauser, Aspect, and Anton Zeilinger for this line of work, not for the 2015 experiments, which were done by other groups. Correlations between entangled particles are stronger than any local hidden-variable theory can produce. Either information travels faster than light (which would violate relativity and has never been observed) or particles genuinely did not have definite local properties before measurement. Most physicists accept the second option. Under this view, reality at quantum level is not locally predetermined. Outcomes are not revealed by measurement but produced by it. Bell’s result does not tell you which interpretation is correct. It tells you that local hidden variables are ruled out – no theory where particles carry predetermined values and only interact locally can reproduce quantum predictions. Nonlocal hidden-variable theories, like pilot wave theory, are not ruled out by Bell’s theorem, but they require influences that propagate instantaneously. Whatever is going on, it is not the gloves-in-boxes story.

One word in that paragraph deserves its footnote. "Loophole-free" means the two escape routes that mattered were closed at once: detectors fast enough that no signal could cross between them in time, and efficient enough that the detected events are not an unrepresentative sample. A third assumption remains, and it cannot be closed even in principle. It is that the two settings were chosen freely, rather than fixed in advance by some common cause reaching back to the source and to both switches. What experimenters can do is push that common cause further away. In 2018 the Big Bell Test let about a hundred thousand people supply the settings by hand. The same year, a group set the detectors using light from two distant quasars, emitted billions of years before the Earth existed. Any conspiracy arranging that violation would have had to be in place at least 7.8 billion years ago. That rules it out of 96 percent of the spacetime volume in the experiment’s past.

This does not allow faster-than-light communication. Entanglement correlations are only visible when you compare measurements from both sides after the fact, using a classical channel. Each individual measurement looks completely random on its own. Only when both datasets are brought together do correlations appear. Nature allows nonlocal correlations but forbids nonlocal signaling. Information still respects speed of light. Causality survives. But the picture of reality that emerges is deeply strange: particles separated by any distance can share properties that did not exist until one of them was measured.

The Counting Argument

Bell’s conclusion is quoted far more often than his reasoning, which is a shame, because the reasoning needs no mathematics beyond counting. Here is a version David Mermin published in 1981 for readers who know no quantum theory at all. Picture two detectors, far apart. Each has a switch with three positions and two lamps, one red and one green. A source between them sends one particle each way. Both switches are set at random, independently, just before the particles arrive. Then one lamp flashes on each side.

Run it many times and two facts stand out. Whenever the two switches happen to be in the same position, the two lamps always match. Every run, without exception. And across all runs, with switches set at random, the lamps match half the time. Both facts are what quantum mechanics predicts for entangled particles with the three settings spaced evenly, and both are what real Bell tests measure.

The first fact looks easy to explain. If matched switches always give matched colors, the natural account is that each particle leaves the source carrying instructions: which color to show at position one, at position two, at position three. Both particles carry the same sheet, which is why matched switches agree every time. That is the whole idea behind hidden variables, and it is exactly the gloves-in-boxes story. There are eight possible sheets, from red-red-red to green-green-green.

Now count. Take a sheet whose three entries are all alike, say red-red-red. Whatever the two switches land on, the colors match: nine agreements out of the nine possible switch pairs. Take a mixed sheet, say red-red-green. The pairs that agree are one-one, one-two, two-one, two-two, and three-three. That is five out of nine. Every sheet is one of these two kinds, so every sheet agrees at least five times in nine. Blend the sheets in any proportion you like, weight them however you please, and the average can only sit at or above five in nine. That is 55.6 percent.

Measurement gives 50 percent. That is under the floor. No choice of instructions, no cunning mixture, no bias hidden in the source can reach it, because the floor came from counting rather than from any assumption about what the particles are made of or how they work. One premise has to go, and only one is available: the particles are not carrying answers for switch positions that were never used.

It is worth being careful about what this does and does not establish, because it is easy to over-claim in both directions. It does not show that something travels between the detectors. It shows that no account in which each particle carries its own local answer sheet can produce the observed numbers. Give up locality instead, and definite properties can be kept, which is precisely the bargain pilot wave theory strikes. Give up the answer sheets instead, and what survives is the split drawn a moment ago: correlations no local story can account for, and still no way to signal. What no theory can do is keep both.

Trying to Send a Message

That last paragraph is easy to nod along to and hard to actually believe, so let us try to break it. Here is the experiment many people reach for when they first meet entanglement. Take an entangled pair and keep one particle while a friend carries the other light-years away – you on Earth, your friend by a distant star, the light between you a decade long. You want to send that friend a single bit right now, faster than any signal could crawl across the gap. And you seem to hold the perfect tool: the instant you measure your particle, theirs is determined, no matter the distance. So the plan writes itself – measure your particle to send a 1, leave it untouched to send a 0. Your friend watches their particle, waiting for the answer. Does it work?

It does not, and the first reason is almost deflating. Your friend, looking only at their own particle, has no way to tell whether you measured yours. They measure it and get an outcome – up or down – but it is a fair coin either way. You measuring first does not tip them toward up or toward down; it only means that if the two of you later compare records, those coin flips will turn out to have been linked. A single result on their side carries nothing. It looks like noise because it is noise. The correlation is genuine, but it lives in the relationship between the two records, and your friend is holding just one of them.

Compare that with the thread at the top of the page. Nothing runs between the two columns here, at any point, because nothing runs between the two particles either. Each side holds a list of coin flips and can stare at it forever without finding anything. The structure exists, and it is perfectly rigid – every pair opposite, no exceptions – but it is a relationship between two lists, and a relationship is not a thing sitting in the space between them. This is also why the correlation cannot carry a message: to see it at all, somebody has to physically bring the two lists together, and that journey is bounded by the speed of causality like everything else.

A sharp reader pushes harder. Fine – drop the measure-or-not trick. Encode the bit in how you measure instead: the vertical axis for a 0, a tilted diagonal axis for a 1. Surely your choice of axis leaves some fingerprint on their side, some shift in their odds they could read off? This is the real attempt, and the answer is the whole point: no. Work it through and their particle behaves identically whatever axis you pick – and identically to the case where you never touch yours at all. On their own, in any basis they choose to measure, they find a flat 50/50, pure static. Your choice is simply absent from their local statistics. This is not a lucky cancellation or a near-miss to be engineered away. It is a theorem: nothing you can do to your half – measure it, rotate it, ignore it – shifts the odds of anything they measure on theirs.

Then what was that "instant" change to their particle? It was a change to the bookkeeping – the single shared description the pair are written into – not a push that arrives at their end and that they could detect or ride. The correlation only steps into view when your record and theirs are set side by side, flip against flip, and setting them side by side means physically carrying one record to the other. That carrying is bound by the speed of causality – the cosmic limit that light merely happens to travel at. Nature offers correlations stronger than any classical story can explain and, in the same breath, forbids you from using them to move a single bit. The two halves feel joined, and they are – but the join transmits nothing.

And the verdict survives no matter which story about measurement you prefer. Say your measurement collapsed the state, or split the world into branches, or only updated what could be known – the observable facts do not budge: their end holds static until the records meet. Even "who measured first" has no settled answer, since relativity lets different observers disagree on the order of two distant events – and it never matters, precisely because nothing is sent. So when you read that measuring one particle "instantly affects" its partner, hold the word lightly. What is instant is an update to a shared description on paper. What is real, local, and bounded by the speed of causality is everything you could ever actually do with it.

So what can you do with it? Not nothing – only not the thing most people first want. Picture it at scale: two civilizations meet, manufacture a million entangled pairs, split every pair, and part across the galaxy. On a schedule fixed in advance, each measures its half of the next pair in the agreed way. Neither side can choose the result; it comes out random. But the two results are locked together by the way the pair was built, so each side, on seeing its own bit, knows the other’s exactly – with nothing ever sent between them. They have not communicated. What they hold is a shared string of random bits, identical on both ends, that they can act on in lockstep across any distance.

An engineer might object: you could get shared random bits the cheap way – print a million coin-flips and take a copy each. Pure coordination needs no quantum mechanics. So what does entanglement actually add? Not a message, but two real things. The first is a secrecy you can certify – by detection, not by magic. There is no pre-written list to photograph, since the values do not exist until someone measures; but a spy who seizes the suitcase can simply measure every particle, and nothing about holding entangled pairs stops that. What stops the spy getting away with it is the protocol. The two sides do not just use the bits – they read each pair in a randomly chosen direction and then sacrifice a random sample, checking in the open that the tell-tale quantum correlations, a Bell-inequality violation, are still intact. A spy who measured first could not have known which direction was coming; every wrong guess collapses a pair and dents those correlations, so the intrusion surfaces as errors and the key is thrown away. Drop that test – fix one agreed direction and just act on the results, exactly as the schedule above does – and a spy with access really can copy the lot and leave no trace. The security lives in the protocol, not in the pairs.

The second thing entanglement adds is strength. The correlations are stronger than any pre-shared list can imitate. For a handful of specific coordination problems, that lets two entangled sides win where no classical agreement, however clever, ever could. Real value, then: certifiable secrets, and a sliver of coordination nothing classical can match. But still, stubbornly, no way to send a chosen bit one inch faster than the speed of causality.

Photons That Never Met

One picture is still standing, and it is the most stubborn one: that entanglement is some kind of thread strung between two particles. There is an experiment that takes it apart. Start with two entangled pairs made independently of each other, photons 1 and 2 from one event, photons 3 and 4 from another. Photon 1 and photon 4 have no shared origin and have never been near one another. Now take the two inner photons, 2 and 3, and measure them jointly, in a way that asks only which combined state the two of them are in and refuses to ask anything about either one alone. Afterwards, photons 1 and 4 are entangled. The procedure is called entanglement swapping, and it was proposed in 1993 by Marek Zukowski, Anton Zeilinger, Michael Horne, and Artur Ekert.

That is already odd. In 2013 a group at the Hebrew University of Jerusalem arranged it so it could not be shrugged off. They separated the two pairs in time. Photons 1 and 2 were created together, and photon 1 was measured at once, which destroyed it. Photon 2 was sent into a delay line of 31.6 meters of open air, long enough to hold it for 105 nanoseconds, the interval between eight pulses of the pump laser. During that wait, a later pulse created photons 3 and 4. Photon 2 came out of the delay and was measured jointly with photon 3. Photon 4 was measured last of all.

Photons 1 and 4 came out entangled. The team reconstructed the joint state of the pair from the measurement records and found it matched the expected entangled state to 77 percent, comfortably past the halfway mark that no unentangled pair can beat. And photon 1 had been detected before photon 4 existed. The two never coexisted. At no instant of the experiment were both of them in the world at the same time.

i Entangled, and there is no instant that contains them both ×

The first objection any careful reader raises is that the two were secretly linked from the start. The team tested exactly that. Spoil the joint measurement in the middle, by delaying one inner photon enough that the two can be told apart, and the swap fails. Photons 1 and 4 then show ordinary classical correlation and no entanglement at all. The link is made by the measurement in the middle. It is not sitting there beforehand waiting to be revealed.

So how does a measurement made after photon 1 is gone leave any mark on it? It does not, and this is the part worth slowing down for. Nothing is sent to photon 1, and nothing needs to be. The joint measurement has four possible outcomes, and each outcome sorts its run into one of four groups. Take every run together, ignoring which outcome came up, and the records of photons 1 and 4 hold nothing remarkable at all: an unsorted heap with no entanglement in it. Sort the runs by the inner outcome, and each group on its own shows the correlations of an entangled pair. The entanglement is a property of a sorted collection of records, and the thing doing the sorting is an ordinary classical result that travels forward in time like everything else.

That is why nothing has to reach backward. Photon 1's result was written down and then simply sat there, a number in a file. The label saying which group it belongs to arrives later. A relationship between two records does not require the records to be made at the same moment, any more than two entries in a ledger have to be written on the same day in order to add up. What the experiment removes is the thread. A thread needs two ends, and there is no instant at which both ends of this one exist.

Now the part that must not be skipped, because the result is easy to misread. Nothing was sent into the past. Look at photon 1's detector record on its own and you find random numbers. Look at photon 4's on its own and you find random numbers. Neither record shifts by a hair depending on what was done in the middle, and no result already written down is ever altered. The correlation appears only when all the records are carried to one place and compared, and carrying them is a physical act bound by the speed of causality, exactly as in the message that could not be sent. Causality is intact. No information travels backward. What changes is not the physics of causes but our idea of what a quantum state is.

A companion result cuts the same way from another angle. Asher Peres asked what would happen if the decision whether to entangle a pair were taken only after both of its photons had been measured and destroyed. In 2012 a group in Vienna carried it out, with the choice made by a quantum random number generator after the fact. Whether the earlier records show an entangled pair or merely a correlated one is settled by a decision made when those photons no longer exist. The lesson repeats. A quantum state is not an object lying somewhere, waiting to be disturbed. It is a description of relationships, and some of those relationships hold between events at different times.

Interpretations

All interpretations of quantum mechanics agree on predictions. They disagree on what is actually happening. Copenhagen interpretation, the oldest and most widely taught, says wave function is a tool for calculating probabilities, not a description of physical reality. Measurement causes collapse. Asking what happens between measurements is meaningless. Shut up and calculate, as some physicists put it. This pragmatic approach dominated for decades and remains the default framework in most textbooks.

Many-Worlds interpretation takes the opposite stance. Wave function is real. It never collapses. Every quantum measurement causes universe to split into branches, one for each possible outcome. You see only one result because you are in one branch. Other outcomes happen in other branches, equally real, forever inaccessible. No collapse. No measurement problem. Just an ever-branching tree of parallel realities. The cost is accepting that your reality is an unimaginably thin slice of a vastly larger multiverse.

Decoherence provides a partial answer without picking sides. When a quantum system interacts with its environment, superposition does not vanish but leaks into the surroundings. Information about the quantum state spreads irreversibly into air molecules, photons, detector atoms. For all practical purposes, interference between branches becomes undetectable. System appears to have collapsed into one outcome even though mathematically all branches still exist. Decoherence explains why we do not see cats in superposition without explaining why we see one specific outcome. It narrows the mystery without solving it.

Pilot wave theory, proposed by de Broglie and developed by Bohm, restores determinism. Particles always have definite positions, guided by a real physical wave. No collapse, no branching, just particles surfing a quantum wave that determines their trajectories. Price is nonlocality: pilot wave must respond instantaneously to distant measurements, making it explicitly faster-than-light in its internal mechanism (though it still cannot transmit signals). Each interpretation is internally consistent. Each makes identical predictions. Choosing between them is currently a matter of philosophical preference, not experimental evidence.

With one important exception, and it is worth naming rather than gesturing at. There is a family of proposals – collapse models, the best known due to Ghirardi, Rimini and Weber, and to Diosi and Penrose – which are not interpretations at all. They are different physics: they add a genuine collapse to the equation, rare enough to be invisible for a single particle and fast enough to be certain for a cat. Because they change the equation, they predict things quantum mechanics does not, and the most useful is a faint spontaneous glow, since a randomly jolted charge must radiate. In 2020 a team looked for exactly that glow with a germanium detector under the Gran Sasso mountain, where the rock screens out almost everything else. They did not find it, which killed the natural parameter-free version of the Diosi-Penrose model. Versions with an adjustable length parameter survive, and the search continues. So the boundary between interpretation and theory is not fixed: some of what once sat safely in the philosophy column has been moved into the laboratory and cut.

Quantum Information

Entanglement is not just a curiosity. It is a resource. A quantum bit, or qubit, can exist in superposition of 0 and 1 simultaneously. Two entangled qubits share a single quantum state that encodes correlations no classical system can replicate. With 300 qubits in superposition, number of simultaneous states exceeds number of atoms in the observable universe. This is the foundation of quantum computing – not faster clock speeds, but a fundamentally different logic. The common shorthand, that it simply tries every possibility at once, is misleading: a measurement still yields only one. The power comes from arranging entanglement and interference so that wrong answers cancel and the right one survives – which is why the speedup is real only for particular problems, like factoring and quantum simulation, not for computing in general.

i Quantum processor: superconducting qubits cooled to near absolute zero ×

Quantum teleportation uses entanglement to transfer the quantum state of a particle to another particle at a distant location. The original state is destroyed in the process (no-cloning theorem forbids copying quantum states), but a perfect replica appears elsewhere. No matter travels. No information moves faster than light, because classical communication is still needed to complete the protocol. What teleports is quantum information itself, the exact state of a system, transferred using entanglement as a channel. This has been demonstrated with photons across hundreds of kilometers and between orbiting satellites and ground stations.

Quantum cryptography exploits a different feature: an eavesdropper who measures a particle in the wrong basis disturbs the correlations, and the disturbance surfaces as errors. Two parties can therefore build a shared key and then check, in the open, whether anyone was listening. What this removes is the assumption that some mathematical problem is hard to solve, which is what ordinary cryptography rests on. What it does not remove is everything else. The guarantee belongs to the idealized protocol; real hardware leaks through side channels, and the classical conversation used to compare notes still has to be authenticated some other way. The security lives in the procedure, not in the particles. This technology is deployed in limited commercial networks and government links.

Deeper still, some theoretical work suggests entanglement may be connected to structure of spacetime itself. In 2013 Juan Maldacena and Leonard Susskind pointed out that for a particular pair of model black holes, the entangled state and a wormhole joining them are two descriptions of one situation. They went on to conjecture that the correspondence reaches all the way down to ordinary entangled particles. That reach is a conjecture rather than a result, and it is worked out only in model universes rather than ours. Any such picture also has to account for pairs that were never in existence at the same moment, which is a stiff requirement for anything shaped like a bridge. But if something along these lines holds, understanding entanglement may be essential to understanding what space and time actually are. Quantum information is not a branch of technology. It may be a branch of fundamental physics.

What Is Real

At its core, entanglement forces a question that physics has never fully answered. Is the wave function a real physical thing, like a field, existing out there in the world? Or is it merely a mathematical tool, a bookkeeping device that tracks what we know? This question, psi-ontic versus psi-epistemic, sounds abstract but has concrete consequences. If wave function is real, then superposition, entanglement, and everything quantum mechanics describes is physically happening. Reality is genuinely nonlocal, genuinely indeterminate, genuinely strange. If wave function is just information, then something else underlies it, something we have not found.

i Is the wave function real, or just a map of our ignorance? ×

There is also a matter of scale that changes how the whole subject reads. Entanglement is usually presented as a delicate effect coaxed out of a laboratory, and that is backwards. Among all the states two systems could be in, the ones that factor into independent halves are a vanishingly thin sliver; take a state at random and it is entangled. Nor is this confined to careful apparatus. The ground state of almost any piece of ordinary matter is entangled, and several everyday phenomena are entanglement wearing other names: the covalent bond holding a molecule together, the alignment of spins in a magnet, the paired electrons carrying a supercurrent. What actually needs explaining is not why entanglement happens but why we so rarely notice it – and that is decoherence, the leaking of the correlations into a vast environment where no one can gather the records back up. The strange, hard-won thing is not the entangled state. It is the ordinary one.

One last number is worth carrying away, because it turns the usual moral on its head. Bell’s inequality gives a ceiling of 2 for any local classical account. Quantum mechanics beats it, but not by as much as it could: no quantum state whatsoever can push that quantity past about 2.83, a limit found by Boris Tsirelson. And yet correlations far stronger are perfectly conceivable without breaking anything. Sandu Popescu and Daniel Rohrlich described hypothetical devices that reach the absolute logical maximum of 4 while still forbidding any signal, so relativity would have no complaint about them. Nature declines to use them. Quantum mechanics sits in a narrow band, stronger than any classical story and strictly weaker than logic permits, and nobody knows what principle puts it there. The usual telling of Bell’s theorem is that the world is stranger than we thought. The full telling is that it is stranger than classical physics allows and tamer than it had to be, and the second half of that sentence is the part still waiting for an explanation.

Recent no-go theorems, most notably the PBR theorem from 2012, have shown that certain classes of psi-epistemic models are inconsistent with quantum predictions. They narrow the space for "wave function as mere information" interpretations without fully closing it. Experiments keep pushing boundaries, testing quantum mechanics in regimes its creators never imagined. So far, not a single prediction has failed. Whatever wave function is, it works with a precision unmatched by any other theory in the history of science. Entanglement sits at the heart of this mystery, connecting particles, challenging locality, and suggesting that nature is stranger and more interconnected than any classical intuition allows.