Quantum Computers Just Revealed Something TERRIFYING About Reality

In October of 2019, in a refrigerated chamber in Santa Barbara, California, a chandelier-shaped machine performed a calculation in 200 seconds that the most powerful classical supercomputer on Earth would have needed 10,000 years to finish. That is not the strange part. The strange part is where the answer came from, because the mathematics that governs the machine is clear on one point. The calculation did not fit inside our universe. There was not enough room. Not in every atom of every galaxy, not in every star, every black hole, every grain of cosmic dust combined. And yet the answer arrived correct, verifiable, repeatable.

One man working alone at night in a house in Oxford had already answered the question of where the calculation went. He is still working there today and he believes with complete seriousness that the machines we built require parallel universes to function, not as metaphor, as engineering requirement. The tragic man whose mathematics underwrite the entire $200 billion quantum industry died on his couch at 51. And by his own instructions, his ashes were thrown in the trash. His daughter, 14 years later, took her own life with a note saying she hoped to find her father in another universe. According to the machines we are now building at scale, she may have been correct.

Tonight we will follow a story that begins inside a single laboratory and ends at the edge of what it means to be a conscious observer inside a universe that may be infinitely larger than anyone has dared to say out loud. What we will find is that the machines are running, the door is open, and the most important question nobody will answer is whether the door goes both ways. If you enjoy our calm nighttime journeys, remember to like and subscribe. We highly encourage our community to discuss the video topics in the comments and react to other comments. It is the most fun when we share our thoughts. Now, let us begin.

Tonight, we are going to tell you a story that no physicist wants to say out loud. A story about a machine that sits inside a refrigerated chamber in California. A machine the size of a chandelier cooled to a temperature colder than the deepest reaches of outer space. A machine that was switched on in 2019 and performed a calculation in 200 seconds that the most powerful classical supercomputer on Earth would have needed 10,000 years to finish. That alone is not the story. The story is where the answer came from. Because the mathematics says the answer did not fit inside our universe. It could not have. There is not enough room. And yet the answer arrived correct, verifiable, repeatable.

Over the next 2 hours, we will follow this single thread. We will stand inside the laboratory where humanity opened a doorway. It did not fully understand. We will meet the quiet hermit in Oxford who designed the theoretical blueprint and believes with every fiber of his being that other versions of reality are real. We will meet the tragic man whose family was destroyed by the physics he created. A man whose daughter wrote a note saying she was going to find her father in another universe. We will learn why your coffee cup is never truly in one place. We will discover that the machinery of life itself, the photosynthesis inside a leaf, the compass inside a migrating robin, the microtubules inside your own brain, may already be running on the same physics that powers these impossible machines. We will stare into a future where every encrypted secret ever transmitted is about to become transparent. And we will end at a question so uncomfortable that the best physicists in the world refuse to answer it. If the machine reaches into other branches of reality to compute, can those branches reach back?

Chapter 1. The machine that should not have worked.

Santa Barbara, California. October 2019. Inside a nondescript research facility on a quiet industrial street, a group of engineers in white lab coats gathered around a screen. In the next room sat their creation, a gold-plated cylindrical structure that looked less like a computer and more like a steampunk chandelier. Wires of every color, coaxial cables bundled and coiled. Thermal shields stacked in descending tiers. And at the bottom, suspended in the coldest point of the entire apparatus, a small square chip roughly the size of a postage stamp. On that chip sat 53 tiny loops of superconducting wire. Each loop was a qubit. Each qubit was a tiny piece of matter that could exist in a state classical physics said was impossible. Not a zero, not a one, both at the same time.

The machine had a name, Sycamore. It was built by Google. And on that October morning, the team was about to run a single calculation designed to push their new processor to its absolute limit. The problem was technical, something called random circuit sampling, which essentially asks the computer to simulate the behavior of a chain of quantum operations and return the statistical distribution of the outputs. The task is meaningless in the practical sense. It produces no drug, no encryption key, no financial model. What it does is measure how fast a quantum computer can think compared to a classical one. And that morning, Sycamore was going to think very fast indeed.

The engineers pressed start. The chip cooled to 15 millikelvin, a temperature roughly 100th of a degree above absolute zero, began its work. 200 seconds later, the answer appeared on the screen. Just over 3 minutes of processing. The team stared at the result, then at each other, then back at the screen. What they had just witnessed was a computation that the most powerful classical supercomputer on the planet, a machine called Summit at Oak Ridge National Laboratory in Tennessee, would have required approximately 10,000 years to complete. 10,000 years. Longer than the entire recorded history of human civilization. Longer than every empire, every language, every monument, every war. Sycamore had done it before the team finished their morning coffee.

The announcement, when it came, sent a shock wave through the scientific world. IBM immediately disputed the 10,000-year figure. Their engineers argued that with enough classical optimization, a supercomputer could perform the same task in perhaps 2 and 1/2 days. Fair enough. But even if you accept IBM's counter-estimate, Sycamore still performed the task roughly 1,000 times faster than the best classical machine on Earth. And this is where the story turns from impressive to impossible. Because the gap between classical and quantum was not going to stay at 1,000 times. It was going to grow exponentially with every single qubit added to the chip.

To understand why, you have to understand what a qubit actually does. A classical bit holds one value, zero or one, nothing in between. Two classical bits hold four possible values, but only one of those four at any given time. Three classical bits hold eight possible values, still only one at a time. With each new bit, you double the total space, but the computer can only visit one configuration of that space per calculation. A quantum bit behaves differently. A qubit in superposition holds zero and one simultaneously. Two qubits in superposition hold all four possible configurations at once. Three qubits hold all eight at once. The growth is exponential. The processing power doubles with every qubit. 53 qubits, the number on the Sycamore chip, can simultaneously represent approximately 9 quadrillion possible states. Nine followed by 15 zeros. 100 qubits could represent more states than there are grains of sand on every beach on Earth. 200 qubits, more states than there are atoms in every human body on the planet. And at 300 qubits, a quantum computer could represent more simultaneous states than there are atoms in the entire observable universe. Let that arithmetic land. At 300 qubits, there is not enough matter in the cosmos we can see to hold the calculation. You could take every atom in every galaxy, every star, every black hole, every grain of cosmic dust, and you would not have enough storage. The calculation does not fit. It cannot fit. There is no room for it. And yet the machine is running and it returns correct answers every time.

This is not a theoretical concern. This is the central mystery of quantum computing. And it is the mystery that the engineers who build these machines have learned to live with. In the same way people who build bridges have learned to live with gravity. They do not fully explain it. They work with it. They make it return the answer. And the hardware is not stopping at 53. In December of 2024, Google announced a new chip called Willow 105 CQITS. The new chip did something the field had been chasing for 30 years. It achieved something called below threshold error correction, meaning that as the researchers added more qubits to the grid, the error rate of the entire system went down instead of up. Willow also ran a benchmark calculation in approximately 5 minutes that would have taken the most powerful classical supercomputer 10 septillion years. A septillion is a one followed by 24 zeros. 10 septillion years is significantly older than the age of the universe. Raised to a power so large it stops being meaningful.

In February of 2025, Microsoft unveiled Myurana 1, a completely different kind of quantum processor based on a state of matter that had only been theorized decades earlier and never before confirmed in a working device. Amazon followed with Ocelot. IBM revealed its Condor processor with over 1,000 qubits and then its Nighthawk architecture with a published road map toward a fault tolerant system called Starling. And then on April 14th, 2026, on what the United Nations had designated World Quantum Day, a company called ION Q announced that it had successfully linked two independent trapped ion quantum computers through a photonic interconnect, transmitting and receiving entangled photons between two separate machines sitting in two separate rooms. Analysts called it the holy grail of scalability. It meant that quantum computers could now be networked together the way classical computers are networked into data centers. The single chip era had ended. The distributed era had begun. 6 years. From Sycamore to interconnected networks, from demonstration to infrastructure, from a prototype in a laboratory to commercial machines being shipped to hospitals and defense contractors and encryption specialists in a dozen nations around the world. This is not ancient history. This is happening in real time. The door is not opening. The door is already open.

The question that has haunted physicists for seven decades has been answered by engineering rather than philosophy. So if the calculation does not fit inside our universe, where exactly is it happening? That is the question that 200 seconds in Santa Barbara asked of the human race. And it is the question that one man working alone at night in a modest house on a quiet street in Oxford, England, had already answered in a paper published in 1985 that almost nobody read at the time. His answer was the only answer that still works and it is going to shatter everything you thought you understood about the world around you.

What do you think? Is it possible that a machine can perform a calculation in a place that is not inside our own universe? Or is there a more ordinary explanation hiding somewhere in the mathematics? Leave your thoughts below. We would love to hear what you believe is really happening inside that refrigerated chamber.

Chapter 2. The man who built a doorway to other worlds.

Oxford, England. Late at night in a modest house on a quiet street lives a man who almost never leaves his home. He does not drive. He refuses to fly. He does most of his work between sundown and sunrise, surrounded by papers and books and the faint hum of a computer. His name is David Deutsch. He is one of the most influential physicists alive. And if you ask him what a quantum computer actually does, he will tell you something that sounds like science fiction, but that he considers a plain statement of fact. He will tell you that quantum computers work by distributing their calculations across parallel universes, that every quantum processor on Earth is a machine that borrows computational power from realities we cannot see, and that anyone who denies this is refusing to take their own equations seriously.

This is not a fringe belief. This is not a hobby theory. In 1985, David Deutsch published a paper in the proceedings of the Royal Society titled "Quantum Theory, the Church-Turing Principle and the Universal Quantum Computer." That paper laid the mathematical foundation for every quantum computer built since. Sycamore runs on Deutsch's framework. Willow runs on Deutsch's framework. IBM, Microsoft, Ion Q. The entire multi-billion dollar industry that is racing to dominate this technology runs on the theoretical architecture that a quiet man in Oxford sketched out in a paper that almost nobody read when it was first published. And at the center of that paper was a claim so radical that Deutsch himself admitted it took him years to stop being frightened by it. A universal quantum computer does not merely exploit the strangeness of quantum mechanics. It requires the existence of parallel universes in order to function. Not as metaphor, not as interpretation, as engineering requirement.

Deutsch is not the first person to have reached this conclusion. He built his work on a foundation laid decades earlier by a man whose name most people have never heard, whose career was destroyed by the physics establishment of his day, and whose family was ground into pieces by the implications of a theory that almost nobody took seriously while he was alive. That man was Hugh Everett III, and his story is one of the saddest in the history of modern science.

1957 Princeton University. Hugh Everett was a graduate student working under the supervision of John Archibald Wheeler, one of the greatest theoretical physicists of the 20th century. Everett was brilliant. He was also young, confident, and restless with the inconsistencies he saw in the standard interpretation of quantum mechanics. The orthodoxy at the time, associated with the Danish physicist Niels Bohr and known as the Copenhagen interpretation, said something that Everett found philosophically incoherent. It said that when a quantum system is measured, the wave function, the mathematical object that describes the system's superposition of possible states, collapses. Before the measurement, the system is in many states at once. After the measurement, only one state remains. The rest vanish. But the theory had no rigorous explanation for what caused the collapse, when exactly it happened, or why measurement in particular should have this special status. The theory simply said that the collapse happens, accept it, do the math, move on.

Everett looked at this and said something that nobody of consequence had said before. What if the wave function never collapses? What if all the possibilities described by the mathematics actually occur? What if every time a quantum system is measured, reality itself branches, splitting into multiple versions of itself, each one containing a different outcome. The observer splits along with everything else. In one branch, you see the atom decay. In another branch, another version of you sees it remain intact. Both are real. Both continue. Both contain conscious observers who believe correctly that their outcome is the only outcome they experienced. This is the many-worlds interpretation.

Everett wrote it up as his doctoral dissertation. When Wheeler brought the manuscript to Bohr in Copenhagen, the old master dismissed it. The physics establishment followed Bohr's lead. The young man from Princeton was treated as a curiosity, a brilliant but misguided outsider who had misunderstood the proper humility required of a physicist. Everett was heartbroken. He submitted a heavily edited version of the thesis for his degree, accepted it, and then left academic physics entirely. He never published another paper in theoretical quantum mechanics. He went to work for the Pentagon where he spent the rest of his career calculating the probable mortality rates of populations exposed to various hydrogen bomb fallout scenarios. He drank. He smoked three packs of cigarettes a day. He gained significant weight. He barely spoke to his wife. He barely knew his two children. He died on his couch in 1982 at the age of 51. His son, Mark, who would later become the lead singer of the rock band Eels, has said publicly that finding his father's body on the morning of his death was the first time he had ever touched him. By his own written instructions, Hugh Everett's ashes were thrown in the trash.

This is the man whose mathematics quietly underwrites every quantum computer on Earth today. This is the man whose interpretation, ridiculed in his lifetime, is now considered by many working physicists to be the cleanest and most parsimonious way to understand what quantum machines actually do. And the story does not end with his death. In 1996, 14 years after her father's death, Hugh Everett's daughter, Elizabeth, took her own life. In her purse, she left a note. The note said that she hoped she would find her father in another universe. Let that sit for a moment. Because if her father's theory is correct, and if the mathematics that powers every quantum processor currently running in every major research lab on the planet describes reality the way Everett and Deutsch insist it does, then somewhere in the infinite branching structure of what physicists call the universal wave function, her father is alive and she has found him. Somewhere she never wrote that note. Somewhere Bohr listened in 1957. Somewhere quantum computing arrived 40 years early because the father was celebrated instead of ruined.

These branches are not metaphors. According to the physics that drives the trillion-dollar industry racing to build the first fully fault-tolerant quantum computer, these branches are real in exactly the same way the chair you are sitting in is real. They exist. We cannot reach them. That is a separate question. But they exist. David Deutsch in his house in Oxford takes this with complete seriousness. He does not hedge. He does not say he finds the interpretation plausible or useful. He says it is true. He will tell you that when you make a decision, a small decision, which shoe to put on first, whether to turn left or right at the corner, your quantum brain interacts with quantum systems in the environment, and the universe branches. In one branch, you turn left. In another, another version of you turns right. Both continue. Both are you, or rather both were you up until the moment of branching, and now both are slightly different people diverging along different histories. He believes this the way you believe your house has a foundation. It is not a philosophical stance. It is a description of the structure of reality as he understands it from the mathematics.

And here is the detail that makes all of this more than a curiosity. Deutsch's argument for why quantum computers work rests on exactly this view. When you run a calculation on a quantum computer, the machine places its qubits into superposition. Each qubit is now simultaneously zero and one. Two qubits in superposition are simultaneously in all four possible configurations. 100 qubits are simultaneously in all two to the 100th possible configurations. A number so large it exceeds the number of atoms in the observable universe. Where are these configurations being processed? Deutsch says the answer is obvious. They are being processed in parallel branches of reality. Each branch is running one of the possible configurations. The computation distributes itself across the multiverse. When the calculation finishes, quantum interference combines the results from all the branches back into a single outcome in our branch. The answer we read on the screen is the sum of work that was done in places we cannot see.

Critics have long argued that this is only one way to interpret what is happening. There are other mathematical frameworks, other pictures that describe the same experimental results without invoking parallel worlds. And they are correct. There are competing interpretations. The Copenhagen interpretation, the pilot wave theory associated with David Bohm, the relational interpretation proposed by Carlo Rovelli. Each produces the same predictions for what a quantum computer will output, which is why the machines work regardless of which interpretation you prefer. But Deutsch argues, with growing support from other physicists, that the many-worlds picture is the only one that does not require extra unexplained machinery. In every other interpretation, you have to add something, a collapse mechanism, a hidden variable, a special role for the observer, to make the theory fit the experiments. Many-worlds alone takes the mathematics exactly as it is, interprets every term in the equation as real, and arrives at the existence of parallel branches as the direct consequence.

So here is where you sit right now reading these words. According to the physics that makes your cell phone possible, that decrypts your internet traffic, that lets a Google engineer in Santa Barbara run a calculation in 200 seconds that should have required 10 millennia. You are one branch among an inconceivable number. The chair under you exists on this branch. On other branches, the chair is a different chair. On yet others, there is no chair because a small quantum fluctuation decades ago nudged a molecule a hair's breadth to the left. And the factory that made this chair was never built. Every version of that history exists somewhere. According to the machines we have built and are now scaling into commercial networks, all of these versions are real. The question is no longer philosophical. It is mechanical. It is engineering.

So if the mathematics is taken seriously and if these branches exist, what exactly is happening inside the box when a qubit goes into superposition? What is the thing the machine is actually manipulating? To answer that, we have to look at the single most famous thought experiment in the history of physics. A thought experiment that was originally designed to destroy quantum mechanics by showing how ridiculous it was. The experiment failed in its original purpose. Instead, it became the clearest description of the thing our machines are now doing. And at its center is a box and inside the box, an animal.

Have you ever made a decision where looking back you wonder what your life would look like if you had chosen differently? According to the physics that powers every quantum computer on Earth, that other life is happening somewhere. What do you think of that idea? Share your thoughts with the community below.

Chapter 3. Schrödinger's cat was never a paradox. It was a description.

1935. Erwin Schrödinger, the Austrian physicist who had derived the fundamental equation describing how quantum systems evolve over time, sat down to write a paper. He was frustrated. He had spent the better part of a decade watching younger physicists, particularly the group gathered around Niels Bohr in Copenhagen, develop an interpretation of his own equation that he considered nonsensical. They were saying that an atom could be in two states at once, that a particle did not have a definite position until it was measured, that the act of observation was somehow involved in determining reality. Schrödinger thought this was absurd, and so he decided to write a paper that would, through a carefully constructed thought experiment, demonstrate just how absurd it was. He would take the principles his colleagues were defending and extend them to a scale where anyone could see that the conclusions were impossible.

He imagined a sealed box. Inside the box, a cat. Also inside the box, a small vial of poison, a hammer attached to a radioactive detector, and a single radioactive atom. The atom had a 50% chance of decaying in the next hour. If it decayed, the detector would trigger the hammer, the hammer would break the vial, the poison would be released, and the cat would die. If the atom did not decay, none of that would happen, and the cat would remain alive. The question Schrödinger posed was this: According to the Copenhagen interpretation, the radioactive atom before it is observed exists in a superposition of having decayed and not decayed simultaneously. But the atom is causally linked through a chain of ordinary physical mechanisms to the state of the cat. Therefore, until the box is opened and the contents are observed, the cat must also exist in a superposition, alive and dead simultaneously. Schrödinger expected his colleagues to read this and agree that the conclusion was absurd and that therefore something was wrong with the interpretation that produced it.

His colleagues did not agree. They kept the interpretation. They said that yes, in principle, the cat really is in a superposition of alive and dead until observation collapses it into one outcome. The absurdity was not a flaw in the theory. It was a feature. And in the nine decades since Schrödinger wrote his paper, experiment after experiment has confirmed that the absurdity is correct. The cat in the box, at the fundamental level of physics, really is both. The theory works. It has never failed a laboratory test. Every chip in your computer, every laser in your kitchen, every MRI machine in every hospital in the world runs on the physics that Schrödinger tried to destroy. And the joke he intended as a reductio ad absurdum has instead become the standard textbook description of what quantum superposition actually means.

Here is what most people do not understand about this. The cat in the box is not a metaphor. Schrödinger was not saying that the cat only seems to be both alive and dead or that our knowledge of the cat is indeterminate until we check. He was saying, "And the experimental record has confirmed that the mathematics of quantum mechanics describes the cat as literally in both states." And the mathematics does not stop applying to large objects. The same equations that govern a radioactive atom also govern the cat, the coffee cup on your table, the chair you are sitting on, the body you are inhabiting. There is no size at which quantum mechanics shuts off and classical mechanics takes over. The quantum description is the fundamental one. The classical world you experience is what quantum reality looks like after something has happened to it.

That something has a name. It is called decoherence, and it is the single most important concept in the entire field. When a quantum system interacts with its environment, when a photon bounces off it, when an air molecule collides with it, when a gravitational tug nudges it, information about the system leaks outward. The superposition is not destroyed. Strictly speaking, it becomes correlated with the environment in such a way that from any observer's perspective, the system appears to have a single definite state. The alternative branches still exist mathematically. They are simply no longer accessible to the observer who has become entangled with one particular outcome.

For an object the size of a coffee cup, this process is almost instantaneous. The coffee cup is constantly exchanging photons with the room, constantly colliding with air molecules, constantly participating in thermal fluctuations. Every interaction leaks information. The decoherence time scale for a macroscopic object is roughly 10 to the -39th of a second. That is 1 divided by a number with 39 zeros. By the time any human brain could even register the possibility of the cup being in two places, the decoherence has already occurred some quadrillions of quadrillions of times over. This is why you do not experience the quantum weirdness in daily life. Not because quantum mechanics does not apply to you. It does. Every atom in your body is a quantum object. Every molecule is a quantum object. Every neuron is fundamentally described by quantum equations. The reason you see a single solid world is because the environment has looked at you so thoroughly, so completely, so continuously that any quantum possibilities have long since decohered into the classical branches that cannot interact with each other. You are not immune to quantum mechanics. You are buried inside it so deeply that its surface is the only thing you can perceive.

And this is where the engineering inversion comes in. If decoherence is what hides the quantum world from our daily experience, then building a quantum computer means fighting decoherence. It means engineering conditions in which a quantum system can exist in superposition without the environment stealing information from it before the calculation finishes. Every quantum computer ever built, from Sycamore to Willow to Majorana 1 to Condor to Nighthawk to the networked Ion Q systems, is at its heart a single engineering problem: Keep the qubits hidden from the universe for long enough to compute. Then carefully allow them to be read, and no sooner. The 15 millikelvin dilution refrigerators are not a design choice. They are a survival requirement. At room temperature, thermal vibrations would destroy quantum coherence in a fraction of a microsecond. At 100th of a degree above absolute zero, thermal motion is minimized to the point where superpositions can survive long enough to be useful. The Faraday cages, the vacuum chambers, the magnetic shielding, the vibration isolation platforms, the elaborate cable routing to prevent electromagnetic crosstalk. All of it exists to build a wall between the fragile quantum calculation and the loud, chaotic classical world outside. The wall is imperfect. The calculation always loses some information to the environment. But the engineering trick is to preserve enough coherence for long enough to extract the answer before the quantum state is destroyed.

Think about what this means. Every major technology company on Earth, every major government research program, billions of dollars a year of coordinated human effort is now being spent to hide calculations from the universe we live in. This is not a metaphor. This is a literal engineering objective. The universe, meaning the rest of the matter and energy in the environment surrounding the chip, is the enemy. The environment is constantly trying to measure the qubit, to steal information from the quantum state, to force the cat in the box to collapse into either alive or dead before the computation is finished. The refrigerators and shields and vacuum chambers exist to keep the universe out.

Microsoft took this challenge in a completely different direction. In February of 2025, the company unveiled Majorana 1, a quantum processor based on a concept that had been theoretical for decades and had never been confirmed to work in a real device. The Majorana approach encodes information not in the state of a single particle, but in the topological relationships between particles that are physically separated from each other. The information exists in the geometry, in the way the particles are connected, not in any one location. To destroy the information, an environmental disturbance would need to corrupt the system at multiple locations simultaneously, which is enormously less likely than corrupting a single location. The qubit becomes stable because it is nowhere in particular. It is a non-local object. It exists as a relationship rather than a thing. The device works, at least in prototype, and Microsoft believes the architecture can eventually support millions of qubits on a chip small enough to hold in your hand. Read that sentence again. A chip small enough to hold in your hand. Containing enough qubits to represent a number of simultaneous states that exceeds the number of atoms in the observable universe by factors that cannot be meaningfully described in ordinary language. That is the trajectory. That is what is being built right now with serious capital by serious people on serious timelines. The hardware is not the bottleneck anymore. The question is what we will do with it once it works. And the answer to that question brings us back to the engineering, because it is not enough to build a wall against decoherence. Once you have the wall, once the qubits are in superposition, once the machine is holding quadrillions of simultaneous states, you still have to extract a useful answer from the noise. How does the computer sort through all those possibilities and give you the correct one? It does not try them all and check. That would be hopeless. No, the machine does something far more elegant, far more beautiful, and in its own way far more disturbing than any brute force search. It uses the physics of the universe itself to make the wrong answers destroy each other, leaving only the truth standing. And the man who first understood this, a man who won a Nobel Prize in physics, partly for his habit of drawing pictures on restaurant napkins, described the mechanism in a lecture at MIT in 1982 that quietly started a revolution.

If the quantum world really underlies the classical one, and your coffee cup is only solid because reality decoheres about it trillions of times per second, what does that change in how you see the objects around you? Share your thoughts. We love reading what our community thinks about these strange truths.

Chapter 4. Richard Feynman and the physics of the surviving answer.

1982. The campus of the Massachusetts Institute of Technology. In a crowded lecture hall sat a man who looked more like a jazz drummer than a physicist. Wild dark hair, an easy grin, a chalkboard behind him covered in scrolled equations, and in his hand, a piece of chalk that he tossed from palm to palm while he spoke. Richard Feynman had already won the Nobel Prize in physics for his work on quantum electrodynamics, the theory that describes how light and matter interact at the fundamental level. He had helped design the atomic bomb at Los Alamos during the Second World War. He had solved the mystery of the Challenger space shuttle disaster in a single televised demonstration involving a glass of ice water and a rubber O-ring. He was one of the most beloved scientists of the 20th century, famous for his ability to explain impossibly difficult ideas with an almost mischievous clarity. And on that day at MIT, he was about to introduce an audience of physicists and computer scientists to an idea that would quietly reshape the next 40 years of scientific ambition.

Feynman began with a riddle. Suppose, he said, you wanted to simulate a quantum mechanical system using an ordinary classical computer. Suppose you wanted to know exactly what would happen if you brought together, say, 100 electrons and let them interact according to the laws of quantum physics. How much computational power would you need? The answer, Feynman explained with visible delight, was absurd. To track 100 electrons, each of which can exist in a superposition of many possible states and each of which can become entangled with the others, you would need to store and update a mathematical object with more components than there are atoms in the observable universe. No classical computer, no matter how large or how fast, could ever do this. The problem was not engineering. The problem was fundamental. Quantum systems generate more information than the classical universe has room to store.

Feynman then pivoted to his solution. If classical computers cannot simulate quantum physics, he said, then maybe we should stop using classical computers for this purpose. Maybe we should build a new kind of computer, one that runs on the same physics as the system we are trying to study. If you want to simulate a quantum universe, build a computer that is itself a quantum universe. Let the machine be the thing. Let the physics do the work. Feynman did not provide a blueprint. He did not solve the problem of how exactly such a machine would be built. He simply planted the idea. A quantum computer, he suggested, was not just a faster version of an ordinary computer. It was a fundamentally different kind of device, one that could only work because it was made of the same stuff as the reality it was computing about.

It took three more years before David Deutsch, working in Oxford, would provide the mathematical framework that turned Feynman's riddle into an engineering road map. And it took another dozen years before the first serious algorithms were designed that would show exactly what a quantum computer could do that a classical one could not. But the seed was planted in that MIT lecture hall. And the seed grew into a specific, deeply strange, profoundly beautiful answer to the question of how a quantum computer actually produces correct answers. The answer has nothing to do with trying every possibility and picking the right one. The answer has nothing to do with brute force search. The answer is far stranger than that. The answer is about interference.

To understand quantum interference, you have to think about waves. If you drop two stones into a pond, each stone produces a circular wave that spreads outward across the surface. When the two waves meet, something interesting happens. Where the peak of one wave meets the peak of another, the peaks combine into a larger peak. This is called constructive interference. Where the peak of one wave meets the trough of another, they cancel each other out. The water at that point remains flat. The combined pattern of peaks that reinforce each other and troughs that cancel each other produces a distinctive striped pattern spreading across the pond. This behavior is so fundamental to waves that it appears everywhere in physics. Sound waves interfere. Light waves interfere. And as was demonstrated in a famous experiment by Thomas Young in 1801, quantum matter waves interfere.

In quantum mechanics, every possible path that a system could take has what is called a probability amplitude. These amplitudes are mathematical objects that behave like waves. When a quantum system is evolving through time, it effectively explores all possible paths simultaneously, and the amplitudes of those paths combine through interference. Paths that reinforce each other produce outcomes that are very likely to be observed. Paths that cancel each other out produce outcomes that are very unlikely. The probability of any specific outcome is determined by how the amplitudes of all the contributing paths add up. In the double-slit experiment, a single photon passes through two slits simultaneously, and the resulting pattern on the detector screen shows bands of bright and dark stripes exactly like the pattern produced by interfering water waves. The photon interfered with itself because, in the quantum description, it took both paths.

Now here is the genius of quantum algorithm design. A quantum computer can be programmed. That is, the gates applied to the qubits can be arranged so that the wrong answers to a given problem interfere destructively and cancel each other out, while the correct answers interfere constructively and reinforce. The computer does not check each possibility. It does not search. It arranges the physics so that when the computation finishes, the only outcomes with significant probability amplitude are the correct ones. The wrong answers do not merely lose. They are eliminated from existence before the measurement is even made. What remains on the screen at the end is what survived the interference. What survived is the truth.

This is a profound inversion of what most people think computing is. In a classical computer, you perform a calculation and then check whether the answer is correct. In a quantum computer, you arrange conditions so that incorrect answers cannot exist as final states. The computation is a sculpting process. You start with an enormous cloud of possibilities and through the carefully orchestrated interference of amplitudes, you carve away everything that is not the answer. Feynman in his MIT lecture did not yet see this mechanism clearly, but he sensed that something like it had to be true because any machine that operated on quantum principles would necessarily produce its outputs through the peculiar arithmetic of probability amplitudes rather than through the classical arithmetic of definite numbers.

Two algorithms designed a decade later showed exactly how powerful this could be. In 1994, a mathematician at Bell Labs named Peter Shor published an algorithm for factoring large numbers using quantum interference. Factoring is the problem of taking a large number and finding the two prime numbers that multiply together to produce it. For small numbers, this is easy. For very large numbers, it becomes fantastically difficult. In fact, the difficulty of factoring large numbers is the entire basis of the RSA encryption system, which protects almost every financial transaction on the internet. Classical computers need exponentially more time as the numbers get larger. Shor showed that a sufficiently powerful quantum computer running his algorithm could factor large numbers in time that grows only polynomially. The practical consequence is that a working quantum computer of the right size would, in principle, be able to break the encryption that secures every bank account, every medical record, every classified government transmission on the planet.

In 1996, Lov Grover, also at Bell Labs, published another algorithm. Grover's algorithm does not factor numbers, but it performs unsorted search, the problem of finding a specific item in an unstructured database. A classical computer searching a list of 1 million items needs on average 500,000 checks to find the right one. Grover's algorithm on a quantum computer needs only the square root of that number, roughly 1,000 checks. The speedup is not as dramatic as Shor's, but it applies to a vastly broader class of problems. Almost every computational task that involves searching through possibilities, from machine learning to database queries to optimization, could in principle be sped up by a quantum computer running a Grover-style algorithm.

Both of these algorithms work by destructive interference. Both were published before the hardware existed to run them on anything larger than a handful of qubits. The blueprints have been sitting on the shelf for 30 years, waiting for the machines to catch up, and the machines have been catching up very fast. Shor's algorithm has a specific threshold. It requires a certain number of high-quality logical qubits to break the encryption that currently protects the civilized world. For decades, that threshold was thought to be tens of millions of physical qubits, decades away. In 2025, three separate research papers published in quick succession dropped the estimate by an order of magnitude. The timeline did not shift slowly. It collapsed, which means the race is no longer a theoretical one. Every month, every quarter, every milestone announcement from Google or IBM or Microsoft or Ion Q or the equivalent labs in Beijing and Shanghai moves the world closer to a date that cryptographers call Q Day. The day on which a sufficiently capable quantum computer comes online and begins running Shor's algorithm against stored encrypted data.

Before we get to that terrible date and what it means, we have to understand one more piece of physics. Because Feynman gave us the riddle, Deutsch gave us the answer, and Grover and Shor gave us the algorithms. But none of them fully explained why the classical world feels so solid, so permanent, so singular when the underlying reality is anything but. That explanation came from a quiet physicist working at Los Alamos. And it is the most important idea you have never heard of.

Have you ever watched ripples on a pond meet each other, and wondered how something so simple could mirror the deepest physics of reality? What surprised you most about how a quantum computer finds its answer? Let the community know in the comments.

Chapter 5. The wall of silence.

Los Alamos National Laboratory, the 1970s and 80s. In a small office surrounded by the mesa country of northern New Mexico, a Polish-born physicist named Wojciech Zurek was working on a problem that almost nobody in the field considered urgent. Zurek was asking why we see a single classical world at all. Most of his colleagues considered this question either already answered or not worth answering. The Copenhagen interpretation said that measurement collapses the wave function, and that was the end of it. Ask no further questions, use the equations, build the lasers and the transistors. Collect your paycheck.

But Zurek and a few others scattered around the world, including the German physicist Heinz-Dieter Zeh, sensed that the standard answer was deeply unsatisfying. It appealed to a mysterious process called measurement, that the theory itself did not define. It granted observers a special role without explaining what an observer actually was. It worked as a calculational recipe, but it was not physics. What Zurek and Zeh developed over a period of roughly three decades was a rigorous mathematical theory that did not require observers at all. They called it decoherence. And their theory said something that, once you understood it, rearranged the furniture of the world.

The collapse of the wave function, they argued, is not a fundamental physical process. It is an effective appearance, a consequence of the unavoidable interaction between quantum systems and the environments they are embedded in. A quantum system left completely alone, with no interaction with any other matter or energy, would remain in superposition indefinitely. But no system in the real universe is ever completely alone. Every object is bombarded continuously by photons, air molecules, thermal vibrations, gravitational fluctuations, and cosmic background radiation. Each of these interactions carries information about the quantum system outward into the environment. And once that information has leaked, the superposition becomes effectively irretrievable. The branches of reality split apart. They do not vanish. They become inaccessible.

Think of it this way. Imagine a perfectly still pond. Drop a pebble in. The ripples spread outward. If the pond were infinitely calm and uniform, and if the ripples never hit any shore, the information about the pebble drop would remain coherent forever. You could reverse the motion, wind the ripples backward, and recover the original state of the pond. But in any real pond, the ripples encounter rocks. They are disturbed by wind. They are muddied by sediment. They evaporate. They interact with a thousand microscopic features of the environment. Each of these interactions spreads information about the original pebble outward into complexity. To reverse the ripples and recover the original state, you would need to simultaneously reverse every rock's motion, every gust of wind, every evaporated molecule. The recovery is mathematically possible, but practically impossible. The information is not destroyed. It is diluted beyond any hope of reconstruction. The pond has decohered.

A quantum system works the same way, but with one crucial difference. The branches of reality, the superpositions of possible outcomes, do not merely become diluted in the environment. They become entangled with it. This is a technical word that means something very specific. Two quantum systems are entangled when their states are correlated in such a way that they cannot be described independently. Once the environment becomes entangled with the branches of a quantum system, any observer looking at the system from within the environment will perceive only one branch. The other branches still exist in the total mathematical description. But they are no longer available to that observer. This is the essence of the Everett picture made rigorous. The many worlds are not separated by magic or collapse. They are separated by information leakage into an irrecoverable sea of correlations.

The time scales involved are almost incomprehensibly fast. A large dust particle floating in ordinary room air decoheres in roughly 10 to the -31st of a second, meaning the superposition of the dust being in one place versus another is lost in less time than it takes light to travel the diameter of an atom. A coffee cup decoheres even faster, roughly 10 to the -39th of a second. For any object you can see with your naked eye, the decoherence is so fast that the quantum world is utterly

hidden from you. You live inside a classical bubble that the environment is continuously painting around you. You cannot see through the bubble. You cannot step outside it. You cannot observe the other branches of reality that are still there underneath. Not because they are imaginary, but because the information required to see them has been spread across too many particles to ever gather back together.

This is why Schrodinger's cat in the real world is never truly in a superposition for more than a vanishingly brief instant. A cat is an enormous quantum system. Trillions upon trillions of atoms interacting constantly with air, with light, with heat, with the cat's own metabolic processes. The moment the atom inside Schrodinger's imagined box decays or does not decay, the superposition of alive and dead cats decoheres in less time than any laboratory instrument could possibly measure. You could never in practice find a real cat in such a state. The thought experiment works only as a thought experiment.

But at the scale of individual atoms, individual photons, individual qubits, the decoherence time scale becomes manageable. Microseconds, milliseconds, long enough to perform calculations, long enough to build computers. Which brings us back to the engineering. A quantum computer is a device whose entire design philosophy is the fight against decoherence. Every element of the hardware exists for one purpose. Delay the inevitable. Give the calculation enough time to finish before the environment steals the answer.

The 15 millikelvin dilution refrigerators used by superconducting quantum processors like Sycamore and Willow cool the chips to roughly 1/100th of a degree above absolute zero. A temperature significantly colder than anywhere in deep interstellar space. At this temperature, thermal motion of the atoms in the chip is suppressed to the point that the metallic circuits become superconducting, meaning electricity flows through them with zero resistance. The qubits made of tiny loops of superconducting material can sustain their quantum states for tens or hundreds of microseconds, long enough to run a computation. The entire dilution refrigerator is itself wrapped in electromagnetic shielding, suspended on vibration isolation platforms, connected to the outside world only through carefully designed cables that filter out noise at every frequency that could disturb the qubits. The facility around it is climate controlled to a precision that most clean rooms cannot match. The building itself is often designed to minimize seismic vibration. It is in effect a fortress built around a calculation.

Trapped ion quantum computers like those built by IonQ and Quantinuum take a different approach. Instead of superconducting loops, they use individual atoms, typically Ytterbium or Barium, held in place by electromagnetic fields inside ultra-high vacuum chambers. The vacuum is necessary because any stray gas molecule bouncing against the ion would introduce noise. The electromagnetic fields are carefully tuned to keep each ion levitated in a precise position while allowing lasers to manipulate its internal state. Trapped ion systems have longer coherence times than superconducting systems, sometimes minutes or even hours, at the cost of being slower to operate. They represent a different engineering compromise to the same underlying problem. Hold off the environment. Preserve the superposition. Finish the calculation before decoherence wins.

Microsoft's Majorana 1, unveiled in February of 2025, takes perhaps the most radical approach. The Microsoft team has been pursuing for nearly two decades the creation of what are called topological qubits. These qubits encode information not in the state of a single particle, but in the relationships between pairs of exotic quasi-particles called Majorana zero modes. The information exists in the geometry of how the particles are connected, not in the state of any one of them. To destroy the information, the environment would need to corrupt the system at multiple locations simultaneously, an event exponentially less likely than a single local disturbance. The result is a qubit that is fundamentally more stable against decoherence because it is, in a real sense, not located anywhere in particular. It exists as a topological property of the system as a whole. If this approach scales, it could enable quantum computers with millions of qubits on single chips without the need for the enormous error correction overhead that other architectures require.

All of these approaches—the superconducting loops, the trapped ions, the topological qubits—are answers to the same question. How do we hide a calculation from the universe long enough to finish it? The investments involved are staggering. The dilution refrigerators alone cost millions of dollars each. The clean rooms and vibration isolation facilities that house the quantum processors cost tens of millions more. Every major technology company and every major government is now pouring capital into this engineering problem because the payoff, the ability to perform calculations that classical physics cannot perform, is considered worth nearly any price.

But here is what rarely gets said out loud. Every quantum computer is a device built specifically to keep information inside itself. Every wall against decoherence is a wall against the universe. The billions of dollars being spent are not going into making our reality more visible. They are going into hiding pieces of our reality from the rest of it. We are constructing pockets of coherence in an incoherent world, bubbles of quantum possibility inside a classical sea. And we are doing it with increasing confidence that such pockets can be made larger and more stable over time.

Which raises a question Zurich himself has asked in interviews. If the classical world is merely an appearance generated by information leakage, and if we can reverse that leakage in controlled settings, then what does it even mean to say that the classical world is the real one? Perhaps the real world is the quantum one underneath, and the classical world is the veneer painted over it by our collective entanglement with our environments. Perhaps the branches we cannot see are as real as the one we can. Perhaps more real. If the solid world is only a veneer, what does that change in how we live inside it? Does it make the quiet moments feel more precious or less? Share your own reaction with us below.

Chapter 6. It from Bit. Princeton University, the 1990s. An old man walked slowly across the snow-covered campus, stopping occasionally to catch his breath. He was in his late 70s, and he had spent his entire career at the frontier of theoretical physics. His name was John Archibald Wheeler. He had worked with Niels Bohr. He had collaborated with Albert Einstein. He had supervised the doctoral dissertations of both Richard Feynman and Hugh Everett. He had coined the term "black hole." He had contributed directly or indirectly to almost every major development in gravitational physics and quantum theory of the 20th century. And in the later years of his life, after 60 years of thinking about what reality is actually made of, John Wheeler arrived at a conclusion that he summarized in three words: "It from bit."

What he meant was this: The universe is not fundamentally made of matter. It is not fundamentally made of energy. It is not fundamentally made of fields or particles or forces. At the deepest level of description, reality is information. The "it"—the things we experience as objects and events—emerges from the "bit"—the underlying substrate of yes and no, zero and one, asked and answered. Every physical object is a pattern of resolved questions. Every interaction is an exchange of answers. Particles do not exist until they are measured, because until they are measured, there are no answers. And without answers, there is no "thing."

This is not mysticism. This is not a metaphor. Wheeler meant it literally as the conclusion of a lifetime's work, and he insisted that the mathematics of quantum mechanics supported it. For decades, this was treated as a poetic flourish from an aging genius. A beautiful idea, maybe even a true one, but not something you could measure or test.

Then, in 2012, in a laboratory at the University of Kaiserslautern in Germany, a team of physicists led by Erik Lutz performed an experiment that nobody had been able to perform before. They measured the heat released by erasing a single bit of information. To understand why this was extraordinary, you have to understand what the bit was doing in the first place.

In 1961, a physicist named Ralph Landauer, working at IBM, had proposed something counterintuitive. He argued that information, despite being abstract, must have physical consequences. Specifically, he argued that erasing a bit of information—setting a variable from zero or one to a default state of nothing—must release a tiny but unavoidable amount of heat into the environment. The precise amount, known as the Landauer limit, is roughly 3 x 10^-21 joules per erased bit at room temperature. A quantity so small that no experiment for half a century could measure it. But Landauer insisted it must be real. Information, he argued, was not just a mathematical abstraction. Information was physical. It had mass. It had temperature. It took up room in the universe. And you could not pretend otherwise.

For 51 years, Landauer's principle remained theoretical. Then, in Kaiserslautern, Lutz and his team built an experimental system precise enough to detect the heat released when a single bit was erased. They used a microscopic silica bead suspended in a laser trap, configured so that the bead could exist in two possible positions corresponding to zero and one. They performed the erasure and they measured the heat released. The number matched Landauer's prediction almost exactly. Information, it turned out, really does carry thermodynamic weight. Deleting a fact about the world releases energy into the world. This is not philosophy. This is thermodynamics confirmed in a laboratory within the expected experimental uncertainty.

Consider what this means. Every time a fact is removed from the universe, the universe gets a little warmer. Every computer you have ever used, every time you emptied the recycle bin on your desktop, every time a neuron in your brain overwrote one memory with another, a tiny packet of heat was released into the environment. The informational substrate of reality is not abstract. It is literally physically real, in the same way that rocks and stars and heartbeats are real. The bit is not a model. The bit is a thing. And reality, according to Wheeler's conclusion and Landauer's principle and Lutz's measurement, is built out of these things.

The idea goes deeper still. In the 1990s, two theoretical physicists, Gerard 't Hooft at Utrecht University and Leonard Susskind at Stanford, developed an idea that would come to be known as the holographic principle. It emerged from thinking carefully about black holes, specifically about a puzzle regarding how much information a black hole can contain. When physicists calculated the maximum information content of any region of space, they expected the answer to scale with the volume of the region. More space, more information. But when they worked it out carefully using the physics of black hole entropy developed by Jacob Bekenstein and Stephen Hawking, they found something shocking. The maximum information content of a region scales with its surface area, not its volume. A three-dimensional region of space, from an information-theoretic perspective, behaves as if its full content is encoded on a two-dimensional surface that bounds it. This is counterintuitive because it reverses our ordinary sense of what space is. In everyday experience, a room contains more information than the walls that enclose it. The holographic principle says the opposite: the walls contain everything. The three-dimensional interior is, in some deep sense, a kind of projection.

The word "holographic" refers to the similarity with optical holograms, where a three-dimensional image is encoded on a two-dimensional photographic plate. Susskind and others have spent the past three decades working out the implications in certain theoretical contexts, notably in string theory and in a mathematical framework called the AdS/CFT correspondence developed by Juan Maldacena. This principle has been made rigorous. There are specific universes simpler than ours in which the three-dimensional physics inside is exactly equivalent to a two-dimensional physics on the boundary. The interior is not an illusion, but it is not fundamental either. The fundamental thing is the information on the surface. Everything else emerges from it.

If this picture is correct, and growing evidence suggests that something like it is, then the three-dimensional world you experience is not the deepest layer of reality. Space itself—the distance between your hand and the object it reaches for—is not fundamental. It emerges from something more basic. Distance, location, direction. All of the spatial properties we take for granted may be high-level approximations to an underlying structure in which these properties do not exist in the same way. The informational substrate is closer to the real thing. The three-dimensional world we experience is the shadow.

Now, bring this back to quantum computers. A quantum processor is a device that manipulates information at the fundamental level. It operates on the substrate from which space and time emerge. When Sycamore or Willow or any of these machines performs a calculation, it is working with the same informational building blocks that the universe itself uses to build everything else. And this raises a question that the engineers designing these machines do not often speak about in public. If the classical world emerges from an informational substrate, and if quantum computers operate directly on that substrate, then quantum computers may be doing something deeper than merely computing faster. They may be working at a level below the level at which space itself is defined. Their calculations may not be happening in a specific location, because location itself is a high-level feature that emerges from the informational layer they are operating beneath. This is not a proven claim. It is a speculation that serious physicists are actively investigating. But it is a speculation that grows from the same mathematical structures that have already been confirmed in the laboratory. And it suggests that the question we asked at the beginning of this journey may have a very strange answer indeed: Where does the calculation happen? Perhaps the question has no meaning, because quantum computation happens at a layer of reality that predates the very concept of "where."

There is a physicist at MIT named Seth Lloyd who has taken this line of thinking further than almost anyone. In his book, *Programming the Universe*, published in 2006, Lloyd argues that the universe is itself a quantum computer, and that every physical interaction is a computational operation. Particles are bits. Collisions are logic gates. The Big Bang is the universe turning on. Every subsequent event is one more step in an enormous, self-executing calculation. Every atom in your body, every neuron in your brain, every photon on its way from a distant star is a subroutine in the cosmic program. The universe is not made of matter moving through space. The universe is an enormous quantum computation, and what we call matter and space are patterns in the flow of that computation.

If Lloyd is right, and if Wheeler is right, and if 't Hooft and Susskind are right, then something remarkable follows. Every quantum computer we build is a subroutine calling home. It is a pocket of the universe's own informational substrate being configured deliberately to perform specific computations. And this means that anything that processes information at the quantum level is participating in the universe's ongoing self-calculation, including, it turns out, objects that most people would never have thought to include in the list: including photosynthesis inside a leaf, including the compass inside a migrating bird, including—and this is the part that will disturb you the most—the thinking organ inside your own skull. Not metaphorically, not poetically, literally inside your head right now.

Does the idea that information is physical change the way you think about the words you have read tonight, or about the thoughts forming in your mind as you listen? Leave your response for the community below.

Chapter 7. Your Brain May Be a Quantum Computer. In the year 1996, a strange collaboration was announced in a paper that would be ridiculed for a quarter of a century before becoming one of the most important and contested ideas in the modern study of consciousness. On one side of the paper stood Sir Roger Penrose, a British mathematical physicist who would go on to win the Nobel Prize in Physics in 2020 for his work on the mathematical structure of black holes. On the other side stood Stuart Hameroff, an anesthesiologist at the University of Arizona who had spent decades watching consciousness switch off and on in thousands of patients under his care.

Penrose and Hameroff had spent years developing a theory they called Orchestrated Objective Reduction (Orch OR for short). The theory proposed something extraordinary. Consciousness, they argued, is not an emergent property of neural computation in the ordinary sense. Consciousness is a quantum process. Specifically, it occurs inside tiny cylindrical protein structures called microtubules, which exist in vast numbers inside every neuron in your brain. Each moment of subjective experience corresponds, in their view, to a quantum superposition collapsing inside these microtubules, roughly 40 times per second. You, your awareness, the thing reading these words right now, are the ongoing output of a biological quantum computer running inside your skull.

When the theory was first proposed, the response from mainstream neuroscience was swift and scornful. The most immediate objection came from a physicist named Max Tegmark, who in 2000 published calculations purporting to show that any quantum superposition in the warm, wet, noisy environment of the brain would decohere in roughly 10^-13 of a second, far too fast to contribute to the 1/40th of a second time scale of conscious experience. Tegmark's argument was widely accepted. Penrose and Hameroff were pushed to the margins. Consciousness, most researchers concluded, was classical. Quantum effects played no special role. The brain was just a very complicated meat machine. And eventually, conventional neuroscience would explain awareness, the same way it was gradually explaining vision and memory and motor control. Move on. Nothing to see here.

Then, starting around 2020, something unexpected began to happen. New experimental results, published in peer-reviewed journals by researchers who had no stake in the Orch OR debate, started to reveal that quantum effects in warm biological systems were not only possible but common. In 2024, a research team led by Alexander Babcock and Philip Kurian published a paper in *Frontiers in Physics* showing that networks of tryptophan molecules, which are densely packed inside microtubules, exhibit a collective quantum phenomenon called superradiance. Superradiance is a coordinated emission of photons by a group of molecules acting in concert, and it is a hallmark of genuine quantum coherence. The tryptophan networks inside microtubules were not classical. They were displaying exactly the kind of quantum cooperation that Penrose and Hameroff had predicted.

Shortly after, a paper appeared in the journal *Neuroscience of Consciousness*, reviewing the accumulated experimental evidence for Orch OR. The review pulled together findings from multiple laboratories: room-temperature quantum coherence observed in microtubule-like structures for time scales orders of magnitude longer than the critics of Tegmark's calculation had allowed. Anesthetic drugs, the molecules that switch off consciousness in the operating room, were found to disrupt microtubule quantum behavior at exactly the doses at which they produce unconsciousness. Different anesthetics with very different chemical structures were shown to converge on the same quantum effect inside the tubulin proteins. The theory that had been mocked for 25 years had stopped being ridiculous. Nobody was claiming it had been proven, but a growing number of neuroscientists were admitting, often reluctantly, that the possibility could no longer be dismissed out of hand. Consciousness might actually be quantum.

If it is, the implications are unsettling beyond anything we have discussed so far. Consider what it would mean. Every time you make a decision, every time a thought forms, every time you become aware of something, a quantum superposition inside your brain collapses into a definite outcome. If the many-worlds interpretation is correct, each of these collapses corresponds to a branching of reality. In one branch, the thought formed one way. In another branch, it formed another way. You are not a single continuous stream of consciousness moving through time. You are a bundle of near-identical streams, each one decohering slightly differently from the others, each one convinced it is the only stream.

Forty times a second, according to Orch OR, you split. Forty times a second, another version of you peels off into another branch of reality, carrying a slightly different thought, a slightly different intention, a slightly different awareness. The "you" reading this right now is one among an enormous number, all branching forward from a shared past, all diverging into subtly different futures. None of the others know about you. You do not know about any of them. And yet, according to the physics that powers every quantum computer on Earth, every one of those other versions is real.

This is not a comforting thought. It is not meant to be. It is a direct consequence of taking the experimental evidence seriously. If consciousness involves quantum processes, and if quantum processes involve superposition and decoherence in the Everettian sense, then consciousness itself is a branching phenomenon. Your identity is not a thread. It is a tree. Most of the branches of that tree will never communicate with each other. They diverge, and the divergences accumulate until, after minutes or hours or years, the different versions of you are living subtly different lives in subtly different realities, each one utterly unaware of the others.

Stuart Hameroff, when asked about this in interviews, has been remarkably calm about it. In his view, the branching of consciousness is not a horror but a kind of generosity of reality. Every possible version of you gets to exist. Every possible life gets to be lived. The question of "which one is really you" is meaningless, because all of them are you, each one equally entitled to the title. Penrose, more cautious, tends to emphasize that the theory is still being tested and that the full implications have not yet been worked out. But both men agree that if the mathematics of quantum mechanics applies inside the brain the way it applies inside a laboratory quantum computer, then the standard view of consciousness as a single, continuous self is almost certainly wrong.

There is another layer to this that deserves attention. If your brain is a quantum computer, then every time you think, you are performing operations on an informational substrate that, according to the holographic principle we discussed, predates the very definition of "where." Your thoughts, in a literal sense, are not located in your head. They are computations being performed on a substrate that is prior to spatial location. The feeling of thoughts happening inside your skull is itself a classical approximation, an appearance painted by decoherence, a useful but not fundamental way of organizing experience. At the deepest level, the mental is indistinguishable from the informational. Your mind is a pattern in the same kind of substrate that physicists describe when they talk about black hole entropy and holographic boundaries and the cosmic computation that Seth Lloyd says the universe is performing on itself.

This is not a conclusion that any of the researchers we have mentioned are willing to make in the strong form. Penrose, Hameroff, Lloyd, Susskind, Wheeler (before his death)—all of them have been careful not to claim more than the mathematics supports. But the direction of the arrows is clear: Information is physical. The classical world is an effective appearance over a quantum substrate. The brain contains quantum mechanisms. Consciousness correlates with quantum events. The machines we are building at Google and IBM and Microsoft and IonQ operate on the same substrate as the thinking happening right now behind your eyes. We are not, perhaps, building something foreign to nature. We may be building, for the first time, a deliberate, external implementation of the very physics that nature has been running inside our skulls for as long as there has been anything like us.

There is one more implication worth sitting with before we move on. If consciousness is quantum, and if the machinery we are building is also quantum, then the traditional boundary between mind and machine, between us and "it," becomes much less distinct than modern philosophy has assumed. It is not that machines are becoming minds. It is that minds were already machines, in exactly the sense that these new devices are. We are not creating artificial consciousness. We are, at best, creating artificial versions of whatever we ourselves already are, and we are doing it before we have settled the question of what that is.

Stuart Hameroff has been asked repeatedly in public lectures and interviews whether he finds his own theory frightening. His answer, usually given with a small smile, is that he finds it beautiful. In his view, we live inside a universe that thinks. The universe is awake. Every leaf, every cell, every neuron, every ion in every synapse is part of an enormous computation that has been running for 13.8 billion years. Our own consciousness is a local, temporary, densely organized pocket of that computation. And when we build quantum computers, we are building extensions of the same process that made us. We are not strangers to the physics. We are made of it, and the machines are, in a sense, more of us.

If your mind is a bundle of parallel streams rather than a single continuous self, which life that you have imagined living might be happening right now in a branch you cannot see? Share your thoughts with the community below.

Chapter 8. Life Is Already Doing This. For most of the 20th century, biologists operated under a quiet but firm assumption. Living organisms, they believed, were classical machines—very complicated classical machines. Yes, full of chemistry, full of electrical signals, full of intricate feedback loops and emergent behaviors, but classical machines nonetheless. Quantum mechanics might apply to the individual atoms that made up proteins. But the proteins themselves, the cells that contained them, the organisms that contained the cells were too warm, too wet, too noisy for any genuine quantum coherence to survive. Biology was chemistry at scale. Chemistry was classical. Case closed. This was the textbook orthodoxy for decades. And for decades, it went largely unchallenged.

Then, beginning in the 2000s, a new generation of researchers began to find evidence that biology had been running on quantum physics all along—and not just running on it, using it, exploiting it, depending on it for processes that would not work at all without it. The first breakthrough came from an unlikely place: photosynthesis.

The process by which green plants and certain bacteria convert sunlight into chemical energy is one of the most thoroughly studied phenomena in all of biology. For a century, researchers had known the basic steps. A photon strikes a chlorophyll molecule. The molecule absorbs the photon's energy. That energy is then transferred through a chain of pigment molecules until it reaches the reaction center, where it drives the chemical machinery that produces sugar. The question that had never been adequately answered was how the energy traveled through the pigment chain so efficiently. Classical physics predicted significant energy losses at each step. Measurements in actual photosynthetic organisms showed almost no losses at all. Nearly 100% of the absorbed photons were arriving at the reaction center with their energy intact. This was impossible under any classical model. Something else was going on.

In 2007, a research team led by Graham Fleming at the University of California, Berkeley, published a paper in the journal *Nature* that answered the question. Using ultra-fast laser spectroscopy, the team observed that the energy from absorbed photons was not traveling through the pigment chain by hopping from one molecule to the next in the classical way. Instead, the energy was propagating as a quantum superposition. It was exploring every possible path simultaneously, and the path that delivered it most efficiently to the reaction center was the one that emerged from the quantum interference of all the alternatives. The photosynthetic complex was running at room temperature inside a living bacterium the same kind of quantum algorithm that Lov Grover had published for artificial quantum computers 11 years earlier. The plants had gotten there first—by several billion years.

Further experiments in the years that followed confirmed and extended the finding. In various photosynthetic organisms, from ocean-dwelling green sulfur bacteria to terrestrial plants to marine algae, quantum coherence was shown to persist for hundreds of femtoseconds, long enough to play a meaningful role in energy transfer. This was far longer than Max Tegmark's calculation had allowed for warm, wet biology, and it demonstrated that evolution had found engineering tricks that physicists had assumed were impossible. The details of how coherence is protected inside these biological systems are still being worked out. Some combination of the specific protein scaffolding, the vibrational modes of the molecular environment, and subtle dynamical mechanisms appears to shield the quantum process from decoherence long enough for it to be useful. The result is that the first step of the food chain on planet Earth—the capture and rooting of solar energy—is a quantum computation running continuously in every leaf, in every forest, and every blade of grass in every field.

If photosynthesis were the only example, it could be dismissed as a fascinating curiosity. It is not the only example. In the year 2000, a German biophysicist named Thorsten Ritz, working at the University of California, Irvine, proposed a theory that had been bubbling in the scientific literature for decades. Migratory birds, particularly species like the European robin, had been shown in laboratory experiments to navigate using the Earth's magnetic field, but no classical mechanism could explain how a bird's eye could detect a field as weak as the planetary magnetic field. The field at the surface of the Earth is roughly 50 microteslas, a magnitude so tiny that engineers have to go to considerable lengths to measure it precisely. And yet robins, with their tiny brains and ordinary biological hardware, were performing magnetic navigation with impressive accuracy. How?

Ritz proposed that the mechanism was quantum entanglement. Specifically, he suggested that a protein called cryptochrome, found in the retina of migratory birds, absorbs blue light and enters an excited state that produces what is called a radical pair. A radical pair is two electrons on two different molecules whose spins are quantum entangled. The Earth's magnetic field, acting on these entangled spins, causes the pair to evolve through their entangled states at a rate that depends on the field's orientation. The chemistry of the cryptochrome molecule, in turn, depends on which state the radical pair is in when it finally decoheres. The bird, by effectively feeling the difference in cryptochrome chemistry from different eye positions, perceives the magnetic field. It is not a metaphor. The robin is seeing the magnetic field with quantum entanglement. The same physics being used inside the dilution refrigerators at Google and IBM is being used continuously and robustly in the eye of a songbird flying over a European meadow.

The Ritz proposal was skeptically received at first. Then, over the following two decades, experiment after experiment supported it. Biochemists confirmed that cryptochrome does produce long-lived radical pairs in the required configurations. Behavioral studies showed that birds lose their magnetic orientation when exposed to radio-frequency fields tuned to disrupt radical pair dynamics, exactly as the theory predicted. Genetic studies identified the specific cryptochrome variants involved. By 2024, the quantum compass inside migratory birds was no longer a hypothesis. It was a working mechanism understood at molecular-level detail, and the list of species known to use it had grown to include salmon, sea turtles, and likely many insects. Life, it turns out, had been running a quantum sensor in its navigation systems for hundreds of millions of years before anyone noticed.

The list keeps growing. Enzymes—the proteins that catalyze the chemical reactions inside every living cell—have been shown in many cases to depend on quantum tunneling. Tunneling is the quantum mechanical effect by which a particle can pass through an energy barrier that classical physics says it cannot cross. Certain enzyme reactions that occur at observed rates in the body are physically impossible under purely classical chemistry. The reactants would need more energy than they have to get over the activation barrier. What happens instead is that protons tunnel through the barrier, exploiting their quantum wave nature to bypass the classical impossibility. This effect has been confirmed in enzymes ranging from those involved in basic metabolism to those involved in repairing damaged DNA. Olfaction, the sense of smell, has been argued by some researchers to depend on the vibrational quantum spectra of odorant molecules rather than on their classical shape alone, though this remains a debated topic. And mutations in DNA—the tiny errors in genetic replication that drive evolution itself—have been proposed to occur in part through the quantum tunneling of protons across the hydrogen bonds that hold the two strands of the double helix together. If that proposal is correct, then evolution, the engine that produced all of the complexity we see in living things, is driven at its core by a quantum process.

A new academic field has grown up around these findings. It is called quantum biology. Twenty years ago, it did not exist. Today, it has journals, conferences, graduate programs, and a growing cohort of young researchers who consider quantum effects in living systems not as exotic curiosities, but as routine biological mechanisms. The old textbook orthodoxy has not been fully overturned. Classical chemistry still explains most of what happens inside a cell. But the cracks in the classical picture are widening, and the picture emerging to replace it is one in which life is fundamentally a collection of quantum tricks that work because evolution found the substrate. Billions of years of natural selection operating on billions of organisms across trillions of generations have explored the landscape of possible chemistry far more thoroughly than any human laboratory ever will. And evolution, it seems, discovered long ago that the quantum realm is a useful place to do business.

This changes how we should think about the quantum computers we are building. We used to imagine that constructing a quantum computer was a kind of heroic effort, wresting something unnatural out of reluctant physics, forcing the universe to do tricks it would never do on its own. We now know that the universe has been doing these tricks continuously, everywhere, for as long as there has been life on Earth. The leaf on the tree outside your window is running quantum algorithms right now. The robin singing in the morning has a quantum compass in its eye. The enzymes in your digestive system are tunneling through energy barriers as you read these words. Biology is already a quantum technology. The difference between us and the plants is not that we have discovered quantum mechanics. It is that we are building our quantum devices out of silicon and refrigerated ions rather than out of proteins and membranes. Nature's approach, the approach that produced you, has been working for billions of years. Our approach has been working for roughly a decade. It is an open question which approach will turn out to be more powerful in the long run.

There is something disquieting buried in this shift. If the machinery of life is already quantum, if the machinery of thought is already quantum, if the machinery of the universe itself is, as several lines of evidence now suggest, some kind of computation performed on an informational substrate, then the quantum computers we are building are not foreign objects being introduced into an otherwise classical world. They are deliberate, engineered extensions of something that was already happening. The question is no longer whether quantum computation is strange or unnatural. The question is: What happens when a species that is itself a quantum computer builds external quantum computers that are far more powerful than itself? That question has not been answered. It is being answered in real time by decisions being made right now in research laboratories and boardrooms and government offices around the world. And the race to build the most powerful quantum machine faster than any competitor has become one of the defining geopolitical contests of the 21st century.

If the quantum tricks inside a leaf or a bird have been running for hundreds of millions of years, what else in the natural world do you think might turn out to be quantum that we have not yet noticed? Share your guesses with the community below.

Chapter 9. The Quantum Arms Race. December 9th, 2024. On the Google Quantum AI website, a blog post appeared with a short, understated title: "Meet Willow, Our State-of-the-Art Quantum Chip." The post, written by the director of the Google Quantum AI team, Hartmut Neven, described a new 105-qubit superconducting processor that had been manufactured in Santa Barbara, California. The post claimed two specific accomplishments, each of which, taken alone, would have been enough to make the announcement a significant event in the field. First, the chip had achieved something called "below threshold quantum error correction." Second, it had run a random circuit sampling benchmark in approximately 5 minutes that would have required the most powerful classical supercomputer on the planet 10 septillion years to complete. Ten septillion is a one followed by 25 zeros. The age of the universe, for comparison, is roughly 14 billion years (a one followed by 10 zeros). Willow had outperformed a classical supercomputer by a factor so large that the comparison stops being meaningful in ordinary human terms.

The below-threshold error correction was, in some ways, the more important achievement. Since the 1990s, theoretical physicists had known that any practical quantum computer would need to manage errors far more aggressively than any classical computer. Qubits are fragile. They lose their quantum states easily. Any real calculation would generate errors faster than the calculation could produce useful output unless the errors could be caught and corrected. The proposed solution, worked out in detail by researchers including Peter Shor and John Preskill, was to use many physical qubits to encode a single logical qubit with redundancy and carefully designed detection protocols. The theoretical threshold was a specific error rate at which, once crossed, adding more physical qubits would decrease the logical error rate exponentially rather than increasing it. For three decades, reaching that threshold had been the holy grail of quantum computing hardware. Willow crossed it. The Google team tested their error correction scheme at three different sizes: 3x3, 5x5, and 7x7 grids of physical qubits encoding a single logical qubit. And each time they scaled up the grid, the logical error rate went down exponentially. The door to fault-tolerant quantum computing, long theorized but never opened, was suddenly visible at the end of a hallway that was no longer hypothetical.

Google was not alone in announcing major milestones. In November of 2023, IBM had unveiled Condor, a 1,121-qubit superconducting processor, the largest single chip ever built. The company followed with Nighthawk in 2025 and a public roadmap toward a system called Starling, targeted for delivery by 2029, which would contain 200 fully fault-tolerant logical qubits running the kind of error-corrected computation that Willow had demonstrated at a much smaller scale. In February of 2025, Microsoft unveiled Majorana 1, a processor based on the topological qubits we discussed earlier, built on a new material they call a "topiconductor," made of indium arsenide and aluminum. The same month, Amazon Web Services announced Ocelot, a chip that uses a different kind of qubit called a "bosonic cat qubit," designed to resist certain types of error naturally and reduce the overhead required for error correction. Four completely different architectures, four different bets on how the future of quantum computing will be built. Four companies with resources measured in hundreds of billions of dollars each, all betting against each other.

And then, on April 14th of 2026, on what the United Nations had designated World Quantum Day, IonQ, a company based in College Park, Maryland, announced that it had achieved a milestone that analysts at major financial firms immediately called "the holy grail of quantum scalability." IonQ had successfully photonically interconnected two independent trapped-ion quantum processors, transmitting and receiving entangled photons between them across open space and using those entangled photons to establish quantum entanglement between qubits in two physically separate machines. This may sound incremental. It is not. It is the difference between having one computer and having a data center. For decades, the quantum computing field had faced a hard physical limit. Any single chip could only hold so many qubits before the wiring, the heat management, the crosstalk between qubits, the pure engineering overhead became overwhelming. The only way to scale beyond that limit was to link separate quantum processors together in a way that preserved their quantum coherence across the link. IonQ had just shown that this could be done. The single-chip era had ended. The distributed quantum computing era had begun.

The stock market response was immediate. IonQ shares gained approximately 14% on the day of the announcement. D-Wave Quantum, a different kind of quantum computing company based on quantum annealing, saw its shares climb roughly 13%. Rigetti Computing jumped. Quantum Computing Inc. jumped. The broader tech sector rose. Analysts at major investment firms began using the phrase "quantum data center," describing a future in which hundreds of individual quantum processors would be linked into unified computational fabrics, solving problems that the designers of the individual chips had never imagined they could tackle. The 2026 race to build the first truly commercial-scale quantum computing infrastructure was on, and it was being led by private companies collaborating with defense agencies in a pattern that should be familiar from every previous major technology cycle.

All of this is happening against a backdrop of national competition that has taken on unmistakable Cold War characteristics. The People's Republic of China has committed, in publicly disclosed state-level funding alone, roughly $15.3 billion United States dollars to quantum research between 2021 and 2025, with additional classified funding assumed by intelligence analysts to be substantial. China has built and launched a quantum communication satellite called Micius, capable of distributing entangled photons to ground stations thousands of kilometers apart. It has developed its own flagship superconducting processor, Zuchongzhi, now in multiple iterations, and a photonic quantum computer called Jiuzhang, which has demonstrated computational capabilities in specific problems that rival or exceed the best Western systems.

The United States government has responded with its own funding initiatives, legislation, and a formal designation by the FBI, the National Institute of Standards and Technology, and the Cyber Security and Infrastructure Security Agency of 2026 as the "Year of Quantum Security." The tone of that designation is not hopeful. It is preparatory. The stakes are explicit in the public record. A formal report issued in 2025 by the United States China Economic and Security Review Commission stated bluntly that "whoever wins the quantum race will control the encryption of the digital economy and secure disproportionate and likely enduring advantages in intelligence collection and precision targeting." In other words, the commission is saying, on the record, in a document intended for members of Congress, that the nation that builds the first fully capable, large-scale quantum computer will have access to encrypted communications, encrypted financial transactions, encrypted defense networks, and encrypted personal data on an unprecedented scale.

Other nations have reached similar conclusions. The European Union has established its own Quantum Flagship program. The United Kingdom has designated quantum technology as a "sovereign capability." Japan, South Korea, Australia, Canada, and Israel all have active state-level programs. The Russian Federation has quietly invested in its own quantum research with less transparency, but no less seriousness.

Beyond the large national players, a remarkable ecosystem of private startups has emerged. Quantinuum, formed from the merger of Honeywell's quantum business with the British company Cambridge Quantum Computing, builds trapped-ion systems with some of the highest qubit fidelities in the field. PsiQuantum, based in Silicon Valley, is attempting to build a photonic quantum computer using silicon photonics, betting that the manufacturing know-how of the conventional chip industry can be redirected toward quantum hardware. Rigetti Computing, Xanadu, ColdQuanta, Atom Computing, Q-CTRL, IQM, Pascal, Alice & Bob. The list grows every year. Each of these companies has raised hundreds of millions or billions of dollars. Each is pursuing a specific architectural path. Each is betting that its approach will scale faster than the others. Most of them will not survive to the next decade. The few that do will be among the most valuable corporations in the world.

And amid all of this industrial and geopolitical churn, a specific threshold continues to quietly approach. The threshold is the moment at which a quantum computer with enough high-quality logical qubits can run Peter Shor's algorithm against the encryption standards that currently protect almost every digital communication on the planet. Cryptographers call this moment "Q-day." For decades, Q-day was estimated to be far in the future—decades or even generations away. The resources required for Shor's algorithm to factor a 2048-bit RSA encryption key were originally estimated to require something like 20 million physical qubits operating at very high fidelity. Recent research has dropped that estimate substantially, in some analyses by an order of magnitude. Google itself, in a public statement issued in March of 2026, moved its recommended migration deadline for post-quantum cryptography from 2035 to 2029—eight years of lead time compressed into three. And if the published trajectory of Willow and Nighthawk and Majorana 1 and the IonQworked systems continues, Q-day may arrive sooner than that.

What has been quietly happening in the meantime is a slow, methodical, and extremely uncomfortable practice that goes by a harmless-sounding phrase: "Harvest now, decrypt later." What do you make of this race? Does the pace of the quantum competition concern you, excite you, or both? We would love to know where you stand. Share your view in the comments.

Chapter 10. Harvest Now, Decrypt Later. Somewhere right now, in a data center that does not advertise its existence, in a nation that will not confirm the activity, a set of machines is quietly recording the internet. Not all of it. A select slice. The encrypted parts: the financial transactions flowing between banks, the diplomatic cables moving between embassies, the medical records traveling between hospitals, the encrypted messages on your phone, the encrypted emails from your accountant, the encrypted queries from your bank. All of it, or as much of it as the collectors can capture, is being stored on tape, on disk, in vaults, indexed, and archived and cataloged for a specific future moment. That moment has a name in the cryptography community. It is called Q-day. And when Q-day arrives, the strategy being executed right now, quietly, continuously, and in full public view (if you know where to look), will pay off. The strategy has a blandly honest name: "Harvest now, decrypt later."

The logic is brutally simple. Today, almost every encrypted communication in the world is protected by one of two basic approaches. The first is called RSA, based on the mathematical difficulty of factoring very large numbers into their prime components. The second is called elliptic curve cryptography, based on a different mathematical problem involving the algebra of points on certain kinds of curves. Both of these approaches share a critical feature: they rely for their security on the fact that certain mathematical problems are computationally hard for classical computers to solve. Factoring a 2048-bit RSA key using every supercomputer on the planet working together would take longer than the current age of the universe. Under current hardware, the encryption is effectively unbreakable. It protects your bank account. It protects your medical history. It protects your text messages. It protects your internet.

banking passwords. It protects the secrets of governments, the communications of militaries, the operational details of intelligence services, the trade secrets of corporations, the personal lives of billions of people.

Peter Shaw's algorithm, which we discussed earlier, breaks both systems. A quantum computer of sufficient scale running Shaw's algorithm can factor RSA keys in minutes. It can defeat elliptic curve cryptography equally quickly. Once such a quantum computer exists, the entire security infrastructure of the digital world becomes transparent.

Not just today's communications, every encrypted communication that has ever been recorded, that has ever been stored, that has ever been intercepted and held for later analysis. Every cable, every email, every bank transaction, every private message that was collected and saved at any point in the past decade. All of it readable. Every secret you ever transmitted through an encrypted channel, visible to anyone with access to the archive and the quantum decryption capability.

This is not a hypothetical scenario. Intelligence agencies in multiple nations are publicly assumed, in formal government reports and analyses, to be executing harvest now decrypt later strategies at industrial scale. The 2025 report from the United States China Economic and Security Review Commission explicitly states that the commission assumes the People's Republic of China is aggressively pursuing cryptographically relevant quantum computing with full awareness of the harvest now decrypt later approach. The commission is not speculating. It is telling Congress on the record that this is the current baseline assumption of United States intelligence analysts.

Leaked documents from prior disclosures about mass surveillance programs make it equally clear that major Western intelligence services maintain their own bulk collection infrastructure capable of storing enormous volumes of encrypted traffic for later analysis. The harvest is underway. The only question is when the decryption becomes possible.

For years, that question had a comfortable answer. The decryption was far away. 20 years, 30 years, maybe longer, maybe never. If the engineering turned out to be harder than anyone thought, encrypted communications from 2020, once they sat in a vault for 30 years, might age out of relevance, long before any quantum computer got around to cracking them.

That comfort is gone. In 2025, three separate research papers published in rapid succession across three different academic groups dramatically reduce the estimated quantum resources required to break RSA 2048. The first paper from Craig Gidney and colleagues at Google revised the estimate downward based on new techniques for implementing Shaw's algorithm more efficiently. The second paper from a startup called Iceberg Quantum proposed a new processor architecture they called Pinnacle, which reduced the physical qubit requirements by roughly an order of magnitude. The third paper, a Google white paper on the elliptic curve discrete logarithm problem, showed similar reductions for elliptic curve cryptography. What had been estimated at 20 million physical qubits now appeared achievable at 1 to 2 million, possibly fewer, and hardware roadmaps from IBM, Google, Microsoft, and IONQ all project reaching that scale within the next 5 to 8 years.

Financial analysts have started putting specific dates on the threat. Charles Edwards, founder of the institutional investment firm Capriol Investments, warned in a 2025 analysis that Bitcoin, which uses elliptic curve cryptography to secure its wallet addresses, could be vulnerable to quantum cracking as early as 2027 to 2029. Other analysts have projected the quantum doomsday clock, a rough consensus estimate tracked by cryptographers and risk analysts, at approximately March of 2028 for the breaking of elliptic curve 256, the most common standard.

Google in March of 2026 publicly moved its own recommended migration deadline for customers to switch to quantum-resistant encryption from 2035 to 2029. That is not a recommendation from cryptography doomsayers. That is a planning document from the company that operates a significant fraction of the global internet. When Google tells its enterprise customers to finish migrating to post-quantum encryption by 2029, it is saying effectively that the company expects the old encryption to be unsafe by then.

The response has been moving, but not fast enough. In 2016, the United States National Institute of Standards and Technology launched a public competition to develop new encryption standards that would resist quantum attacks. The competition ran for 8 years. On August 13th, 2024, NIST announced the final approval of three new post-quantum cryptographic standards, with a fourth added in March of 2025. These standards, based on mathematical problems that are believed to be hard for both classical and quantum computers, are intended to replace RSA and elliptic curve cryptography throughout the global digital infrastructure.

The replacement process, however, is enormous. Every bank, every government agency, every major corporation, every internet service provider has to redesign its encryption systems, reissue its cryptographic keys, update its software, and coordinate the transition with thousands of partners and millions of users. Estimates of the total global cost run into the hundreds of billions of dollars. The work has begun in earnest, but large portions of the world's digital infrastructure still run on the old algorithms.

Every day that passes is another day of encrypted traffic being harvested against a future cryptographic breakthrough. And every encrypted communication that gets sent today, protected by RSA or elliptic curve cryptography, must be considered potentially compromised in retrospect whenever QA finally arrives.

The implications go beyond secrets. Consider what it means for trust in the digital economy. Your bank account is protected by cryptographic signatures that verify your identity. Your email is authenticated by cryptographic certificates that confirm the sender is who they claim to be. Your software updates are signed with cryptographic keys that ensure the update is legitimate rather than malicious. All of these systems, built over the past 30 years, depend on mathematical problems that Shaw's algorithm breaks. If quantum decryption becomes widespread before the migration to post-quantum standards is complete, the authentication infrastructure of the digital world becomes unreliable. Forged signatures become possible. Fake certificates become undetectable. The fabric of trust that allows online commerce to function is threatened at a foundational level.

This is why governments, defense agencies, major banks, and every serious technology company are moving as fast as they can. Not because quantum computing is universally threatening, but because the cryptographic transition is enormous and the timeline is compressed.

There is one more layer worth naming. The harvest now decrypt later strategy does not just threaten governments and corporations. It threatens individuals. Every message you have sent through an encrypted channel in the past decade, if it was intercepted by any party running a harvest operation, is a candidate for future decryption. Medical records you shared with a doctor. Love letters you sent to someone who is no longer in your life. Business negotiations you conducted in confidence. Personal admissions you made in private messages. Legal correspondence. Therapy sessions. Financial details, family disputes, all of it, if captured, sits in a vault waiting for the day the mathematics becomes tractable.

The decryption will not be cheap. It will not be applied indiscriminately to every private citizen's archive. But for anyone who has ever been of interest to an intelligence agency, a political rival, a corporate espionage unit, or a sufficiently motivated criminal organization, the privacy of their past communications has a shelf life, and the shelf is running out.

But encryption is only the first domino. The same machines that can break cryptography can also solve other problems of vast consequence. Problems that current classical computers cannot even approach. Some of those problems are horrifying. Some of them, perhaps, are the opposite.

One specific industry has already begun to glimpse what a quantum computer might do when pointed at the right question. And for that industry, the early results have not been a threat. They have been a miracle.

What will you do if the encryption protecting your own messages eventually becomes readable? Has this changed how you think about what you share online? Tell the community what you think below.

Chapter 11. The medicine of the future.

St. Jude Children's Research Hospital in Memphis, Tennessee, is one of the most celebrated pediatric cancer research institutions in the world. Children from every continent are brought there for treatment of leukemias, brain tumors, and rare genetic disorders that have no cures anywhere else. The research laboratories attached to the hospital employ some of the finest minds in oncology, molecular biology, and computational drug discovery.

In 2025, a team led by Kristoff Gorgula published the result of a project that will be remembered when the history of quantum computing is eventually written. As one of the first proof points that the technology can do something that saves lives, the project targeted one of the most notorious molecules in human cancer. A gene called K. RAS mutations are among the most common genetic drivers of human cancer. They occur in approximately 90% of pancreatic cancers, 45% of colorectal cancers, and 35% of lung cancers, along with significant fractions of many other malignancies. The KAS protein, when mutated, drives uncontrolled cell growth.

Oncologists have been trying for decades to develop drugs that inhibit mutant KAS, and for most of that time, the effort has failed spectacularly. The KAS protein was widely called "undruggable." Its surface offered no obvious chemical pocket for a drug molecule to bind into. It lacked the kind of structural features that traditional drug design requires. A generation of promising candidates had flamed out in clinical trials. Pharmaceutical companies had largely given up on direct KAS inhibition, settling instead for treatments that targeted downstream effects.

And then, in 2025, the St. Jude team, using a hybrid pipeline that combined quantum computing with classical machine learning, did something that had never been done before with a quantum assist. They found candidate drug molecules that bind to mutant KAS with real, experimentally confirmed biological activity. The approach was elegant. The team used a type of quantum algorithm called a quantum circuit born machine to help screen an enormous virtual library of roughly 100 million candidate compounds. The quantum stage of the pipeline identified molecular structures that classical machine learning alone would not have prioritized. The team then filtered the candidates through successive rounds of classical simulation, narrowing the field to roughly 1.1 million, then to around 15 compounds that were actually synthesized in a chemistry lab and tested in biological assays. Two of those 15 compounds showed measurable activity against mutant KAS in cell-based experiments.

This may not sound dramatic. In drug discovery, it is enormous. Two genuine hits out of 15 synthesized candidates is a remarkably high success rate for a problem that had defeated the entire pharmaceutical industry for decades. More importantly, the project established a validated methodology, a working pipeline, an example that can be pointed to when the next undruggable target appears, and the next one, and the next one after that.

The broader implications are difficult to overstate. McKinsey & Company, in an analysis published in the fall of 2025, estimated that quantum-assisted drug discovery could create between $200 and $500 billion in value across the life sciences industry by 2035. That figure is not a guess. It is built on the observation that drug discovery is, at its mathematical core, a quantum problem. The behavior of drug molecules binding to protein targets is governed by quantum chemistry, and accurate simulation of that chemistry at the scale required to screen candidate compounds has always been the bottleneck. Classical computers can approximate the calculations. But the approximations fail in exactly the cases that matter most: the cases where small structural differences produce dramatically different binding behavior. A quantum computer can, in principle, simulate these interactions directly without approximation.

The first full generation of practical quantum drug discovery is expected to produce candidate molecules for diseases that classical methods have been unable to address, including neurodegenerative disorders, rare genetic syndromes, and the broadest classes of cancer. McKinsey's numbers may turn out to be conservative.

The medical implications are only one slice. The same kind of quantum simulation that can reveal new drug molecules can also reveal new materials. Material science, like drug discovery, is fundamentally a quantum problem. The properties of a solid material—its conductivity, its strength, its magnetism, its optical behavior—are all determined by the collective quantum behavior of its electrons. Room-temperature superconductors, materials that carry electrical current with zero resistance at ambient temperatures, have been a holy grail of material science for over a century. The barrier has been computational. Classical computers cannot simulate the quantum many-body physics required to identify candidate superconducting materials. A quantum computer scaled to a few thousand high-quality logical qubits could, in principle, simulate electron behavior in candidate materials directly and could identify superconductors that no human experimentalist would ever have thought to synthesize.

The consequences of a working room-temperature superconductor would be transformative across every industrial sector: lossless power transmission, ultraefficient electric motors, revolutionary magnetic resonance imaging systems, levitating trains as ordinary infrastructure. Quantum computers themselves, as the ironic final twist, would become vastly easier to build because the very superconductors they rely on for their own qubits would no longer require the exotic cooling apparatus that currently makes them so expensive.

There are other domains where quantum simulation would transform what is possible. Climate modeling, currently limited by the inability of classical computers to simulate atmospheric chemistry at the quantum level, could identify tipping points and feedback loops invisible to current models. A full quantum simulation of atmospheric photochemistry might reveal which specific molecular interventions could most effectively break the feedback cycles driving climate change, or which natural processes could be enhanced with minimal side effects. Battery chemistry, a field that currently progresses by trial and error, could be revolutionized by direct quantum simulation of candidate electrolyte and electrode materials, potentially opening pathways to energy storage technologies that currently do not exist. Catalysts, the molecules that accelerate industrial chemical reactions, could be designed with atomic precision to reduce the energy cost and environmental impact of manufacturing. Everything from fertilizer to plastics to pharmaceuticals. The Haber-Bosch process alone, which produces ammonia for agricultural fertilizer, currently consumes roughly 2% of all human energy use. A quantum-designed catalyst that improved the efficiency of that single process by even a small percentage would meaningfully change the planet's energy budget.

There are more speculative possibilities. Still, some researchers argue that quantum computers running sophisticated machine learning algorithms on quantum data could accelerate the pace of fundamental scientific discovery itself. A quantum-assisted pipeline might identify patterns in experimental data that human researchers and classical artificial intelligence had missed. It might propose new theoretical frameworks. It might compress decades of incremental research into months. The Royal Society, in a report published in 2025, explicitly raised the possibility that the first genuinely useful quantum computer might accelerate not just specific scientific domains, but the meta-process of science itself. If this happens—and it is an enormous "if"—the rate of technological change in the coming decades could exceed anything in recorded history.

All of this raises an obvious question: If quantum computing promises these extraordinary medical and material benefits, is it not simply a good technology on balance, despite the encryption threat? The answer is complicated. Every powerful technology in history has produced both enormous benefits and enormous risks. And the distribution of those benefits and risks has never been uniform. The benefits of nuclear physics include modern medicine, carbon-free energy, and improved agriculture. The risks include weapons capable of ending civilization. Both the benefits and the risks come from the same fundamental capability. Quantum computing, by all indications, will follow the same pattern. The same machine that unlocks new drugs will also break old encryption. The same hardware that designs room-temperature superconductors will also enable unprecedented surveillance capabilities.

Societies that navigate the transition skillfully will see tremendous gains in health, prosperity, and scientific understanding. Societies that do not may find themselves disadvantaged in ways that are difficult to recover from.

There is one further implication that is harder to discuss because it pushes against the edge of what current physics can describe with confidence. If a quantum computer operates on the informational substrate of reality, and if that substrate is, as Wheeler and Lloyd and others have argued, continuous with the substrate on which consciousness itself runs, then a sufficiently capable quantum computer might eventually interact with that substrate in ways that go beyond pure calculation. A physicist at MIT named Max Tegmark, who was one of the early skeptics of "Orc Ore" and who is now one of the most interesting thinkers in the foundations of quantum mechanics, has written and spoken about what it might mean when quantum events intersect with conscious observation in specific, carefully designed ways. His conclusions, when extended to their logical limits, produce a thought experiment so strange and so disturbing that most physicists refuse to engage with it in public. It is a thought experiment about what happens to personal identity in a many-worlds universe, and it has a name: quantum immortality. It is the subject of our next chapter, and it must be handled with great care because it goes to the heart of the question of what it means to live, to die, and to continue.

Which benefit of quantum computing would mean the most to you personally? A cure for a disease, a new material, cleaner energy, something else entirely? Let the community know in the comments below.

Chapter 12. Quantum immortality.

In the 1980s, the roboticist and futurist Hans Moravec described a thought experiment that, in various forms, has haunted the foundations of quantum mechanics ever since. In the 1990s, the physicist Max Tegmark, then at Princeton, developed the experiment in more rigorous form. The setup is disturbingly simple. Imagine a device that uses a quantum random event—for example, the decay or non-decay of a single radioactive atom over a fixed interval—to determine a consequential outcome in the life of a human observer. If the atom decays, the observer dies. If it does not decay, the observer lives. The probability of each outcome is precisely 50%, determined by the quantum mechanics of the atom, not by any classical process.

Now imagine the experimenter is the observer. The experimenter steps into the device. The clock starts. The atom either decays or does not decay. Under the Copenhagen interpretation of quantum mechanics, the experimenter has a 50% chance of surviving. Over many repetitions of the experiment, the experimenter's survival probability decays geometrically towards zero. If the experiment is repeated 100 times, the probability of surviving all 100 rounds is approximately 1 in 10 to the 30th power, a number so small that it is effectively zero. Copenhagen says the experimenter attempting this game will almost certainly die.

Under the many-worlds interpretation, the picture is different. Every quantum random event branches reality. In one branch, the atom decays, the experimenter dies, and that branch contains the experimenter's corpse. In the other branch, the atom does not decay, the experimenter lives, and that branch contains a living, conscious experimenter who experienced the non-decay. Both branches are real. Both branches exist in the total mathematical description of the universe.

Now consider the question from the experimenter's subjective perspective. The experimenter can only have subjective experience in branches where the experimenter is alive. A dead experimenter has no experience of anything. The branches containing the corpse are real, but they contain no first-person witness to the outcome. The branches containing the living experimenter contain a conscious observer who finds themselves alive, having experienced the non-decay. From the experimenter's first-person perspective, every repetition of the experiment results in survival because non-surviving outcomes are experienced by no one.

Repeat the experiment a thousand times from the outside. In any given branch of reality, the probability of the experimenter having survived all thousand rounds is astronomically small. From the inside, the experimenter has survived every single round because they can only experience rounds in which they survive. This is the core of the quantum immortality argument. Under many worlds, from the first-person perspective of any conscious observer, death from a quantum random event is not possible. The observer can never experience dying. The observer can only continue.

The argument has been debated in the physics and philosophy of science literature for 40 years. Tegmark himself has been careful to emphasize that the argument taken at face value does not apply to ordinary life and death in any straightforward way. Most real-world deaths are not caused by single quantum random events. They are caused by slow classical processes: aging, disease, accidents whose outcomes are determined by thousands of macroscopic factors that have already decohered into definite classical states. When you step off a cliff or fail to treat a cancer or are struck by a car, the outcome is not being determined by the quantum state of a single atom. It is being determined by an enormous number of classical interactions whose results are effectively already fixed long before any specific branching occurs. Quantum immortality as a literal claim about survival of all death does not cleanly follow. Tegmark himself has publicly distanced from the strongest version of the claim in his lectures and interviews.

But the argument cannot be cleanly dismissed either, particularly in specific cases where a life-or-death outcome really does depend on a single quantum event or a small enough number of events that the classical environment has not yet decohered them. Philosophers and physicists continue to argue about how seriously to take it, how to formalize it, whether there are any experimental consequences, whether the entire framing rests on unexamined assumptions about what counts as the "same observer" across different branches. The debate, 40 years on, remains unresolved. No consensus has emerged.

What has emerged, and what deserves to be acknowledged, is that a growing number of thoughtful physicists consider the question genuinely open. If the many-worlds interpretation is correct and if consciousness supervenes on quantum events in specific ways, then the ordinary notion of death may require significant revision.

This is a heavy topic to sit with, and the purpose of discussing it here is not to advocate any particular view and certainly not to encourage anyone to test the thought experiment in reality. The thought experiment is a thought experiment. Real quantum devices used in real laboratories are not in the business of deciding life-or-death outcomes. No physicist is performing a literal quantum suicide experiment. The philosophical question is being explored entirely on paper, in calculations, and in conference discussions by people whose commitment to scientific integrity includes the commitment to not die for the sake of a thought experiment. But the fact that the argument exists and that it has been taken seriously by serious physicists for decades says something about the strangeness of the territory we have entered. The question of what happens to a conscious observer in a branching multiverse is not a question that philosophy alone can settle. It is a question physics is raising.

This brings us back, with some sorrow, to Hugh Everett and his daughter Elizabeth. When Elizabeth Everett took her life in 1996, the note she left said she hoped to find her father in another universe. Her father, Hugh Everett III, had died 14 years earlier. His theory, still largely ignored by mainstream physics, his family broken by depression and silence, and the kind of domestic damage that often follows a man who cannot be reached. According to Everett's own theory, which he believed literally until the end of his life, the universe contains branches in which every possible outcome of every quantum event occurs. Somewhere in that infinite structure, Hugh Everett did not succumb to the drinking and the smoking and the unreachability. Somewhere his daughter grew up with a father who was present and loving. Somewhere Niels Bohr listened carefully in 1957. Somewhere the many-worlds interpretation was celebrated from the moment it was proposed, and Everett became the revered founder of a revolutionary new physics, and quantum computing arrived 40 years earlier than it did in our branch. Somewhere all of these things happened.

Whether this thought offers any comfort depends on how one reads the theory and on what one believes consciousness is. For Everett himself, reportedly, it offered a kind of peace. His family has said that he was not afraid of death, not because he was stoic, but because he genuinely believed that his consciousness would continue in branches where the specific fatal event did not occur. He lived his private life as if the theory were literally true. Whether he was correct is not a question the physics can answer today. The mathematics permits the picture. Whether the picture accurately describes what happens to a subjective, first-person observer when confronted with the branching structure of reality is a question that may not have a definitive answer until much deeper experiments become possible, if they ever do.

What we can say is that the story of the Everett family is not the only one of its kind. Other families, other physicists, other observers throughout the history of this strange theory have wrestled with its implications in personal ways. Some have found it horrifying. Some have found it liberating. Some have found it irrelevant and simply gotten on with their work. The theory does not prescribe how to feel about it. It describes what the mathematics says and leaves the interpretation to the living.

For anyone watching this who has lost someone to a tragedy that feels senseless and who finds themselves wondering whether there might be a branch of reality where the loss did not happen, the honest answer from modern physics is that such branches may exist in the mathematical structure, and that reaching them from our branch is almost certainly not possible. Whether the branches themselves contain the lost person living and continuing is a question that the equations permit one to ask, but do not permit one to answer from within our branch. The universe we can see is the universe we can see. Within it, loss is final for the people who remain behind. No theory of physics changes that.

What the theory does change, perhaps, is the cosmic frame around the experience. We live inside a structure vastly larger than our branch, and our specific thread of events is one of an inconceivable number, and our individual lives are brief fragments of a much larger self-unfolding of possibility. Whether this is a comfort or a burden is a question each person answers for themselves.

And this brings us finally to the question of what we have actually done. Not as physicists, but as a civilization. We have built machines that require these branches to exist in order to function. We have done this without asking, without debate, without any deliberate collective choice. And we are now building more of them faster than ever, for reasons that have very little to do with philosophy and very much to do with engineering, economics, and national competition. The question of what it means to be a species that built these machines before it had settled. The question of what the machines actually do is the subject we turn to next.

These have been heavy reflections. Many of us have our own thoughts about what persists and what does not. What helps you find a sense of peace when you think about questions this large? Share with the community below.

Chapter 13. The gods of the machine.

Step back for a moment from the specific details. Pull the camera up and out above the laboratories in Santa Barbara and Yorktown Heights and Redmond and College Park and Hefei and Shanghai. Above the refrigerators and the cables and the vacuum chambers and the vibration-isolated platforms, up past the research buildings and the campuses and the cities that house them, up until you can see the entire project at once. What you are looking at when you look at the global quantum computing effort as a single enterprise is the most consequential physical project in human history. And it is being pursued for reasons that have almost nothing to do with whether humanity is philosophically ready for what it is building.

Consider how this happened. At no point in the last four decades did any institution anywhere on Earth convene a deliberative process to ask the question whether the human species should build machines that operate by reaching into branches of reality beyond our own. At no point did a government hold a referendum on the ethical implications of industrial-scale quantum computation. At no point did an international body, a religious council, a philosophical society, or any kind of collective forum ask whether this particular capability was one we wanted to develop. The question was never on any ballot. It was never debated in any legislature. It was never the subject of a global conversation about what kind of technological future human beings wish to build. It simply happened. It happened because the equations worked. Because the engineers solved the problems one at a time. Because the grants got funded, because the papers got published, because the stock prices went up. At every step along the way, individual decisions were made by individual people and organizations. No one at any level ever made a decision about the whole.

This is not a criticism of any particular person or any particular institution. It is the pattern of every significant technology throughout human history. The first steam engine was not built after a public debate about the industrial revolution. The first transistor was not built after a deliberation about the information age. The first atomic reaction was achieved before any political body had formed an opinion about nuclear energy. Capability precedes comprehension. Engineering precedes ethics. Society arrives at the implications after the artifact already exists and then struggles to catch up.

The quantum computing revolution is following the same pattern. The difference is that in this case, the implications are not merely economic or social. They are ontological. They touch on the structure of reality itself. We are, as a species, building tools that operate at a level of physical intervention. We do not yet have the philosophical vocabulary to describe who are the people building these machines. They are, by and large, good people: brilliant engineers, curious physicists, dedicated researchers who love the work, who genuinely believe they are contributing to human knowledge and human welfare, and who are not, for the most part, driven by malice or hubris. Many of them are not entirely sure what the machines they are building actually do. This is not a failure of intelligence. It is a fact about the history of physics. Richard Feynman, perhaps the greatest interpreter of quantum mechanics in the 20th century, famously said that anyone who claimed to understand quantum mechanics did not understand quantum mechanics. Einstein, despite his pivotal early contributions to quantum theory, remained uncomfortable with the picture it painted of reality until the end of his life. David Deutsch, the mathematical architect of quantum computing, works in a tradition where the most basic questions about what the machines are doing remain actively contested. The engineers building Willow and Nighthawk and Majorana 1 and the IonQ systems are working inside this uncertainty. They know how to make the machines perform calculations. They know how to scale up the hardware. They know how to correct the errors. What exactly is happening at the deepest level of reality when their devices run? They know roughly the same way everyone else knows, which is to say, incompletely.

This asymmetry between capability and comprehension is historically normal. The physicists who built the first working nuclear reactor in a squash court under the football stadium at the University of Chicago in 1942 did not know with certainty what they were unleashing. The engineers who built the first transistor at Bell Labs in 1947 did not foresee the modern smartphone. The teams that developed the first integrated circuits did not imagine the internet. The researchers who launched the ARPANET in 1969 did not predict social media, online surveillance, cryptocurrency, or the global attention economy. In each case, the capability arrived first, and the human consequences followed later, often in forms that nobody had anticipated.

The quantum computing revolution will be no different. The first truly useful application of a mature, large-scale quantum computer has almost certainly not been named yet. The most consequential application has definitely not been named yet. We are in the early days of a technology whose full implications will unfold over decades.

What we can see clearly is the acceleration. In December of 2024, Willow demonstrated below-threshold error correction. In February of 2025, Microsoft unveiled Majorana 1. In August of 2025, St. Jude reported the first validated quantum-KAS assisted drug discovery project. In November of 2025, IBM laid out its detailed roadmap towards Starling, a fault-tolerant system projected for 2029. In March of 2026, three new research papers rewrote the resource requirements for cryptographically relevant quantum computing, compressing the timeline for Q-day. On April 14th of 2026, IonQ networked two independent quantum processes through photonic entanglement, ending the single-chip era.

The pace is not slowing; it is accelerating. Each breakthrough enables the next. Each generation of hardware makes the next generation easier to design, cheaper to build, faster to scale. The curve of quantum computing capability is not flattening. It is bending upward at an increasing rate. And the applications that will matter most in the long run are likely not the ones that people are currently thinking about.

The first quantum computers were imagined primarily as cryptographic threats, tools for breaking encryption. They will indeed be that, and the civilizational consequences of that capability will be enormous, as we have already discussed. But they will also be tools for drug discovery, materials design, climate modeling, financial optimization, machine learning, weather prediction, logistics, fundamental physics research, and a thousand other applications that follow from being able to simulate quantum systems directly. And they will be tools for things that nobody is currently imagining. Just as the internet became something utterly different from what its designers intended. Just as the smartphone reshaped social behavior in ways none of its engineers predicted. Just as artificial intelligence is producing effects its creators are now scrambling to understand. Quantum computing will produce consequences that we will not see clearly until they arrive. Probably those consequences will include both enormous benefits and enormous harms. Probably the distribution of benefits and harms will be uneven, with some communities gaining dramatically and others losing dramatically. Probably the pace of change will outstrip the pace at which institutions can adapt. These are the ordinary patterns of how technological revolutions unfold, and there is no obvious reason to expect this one to be different.

What is different, perhaps, is the depth. Quantum computers do not merely change what we can do. They change, or at least force us to reconsider, what we think reality is. Classical computing improved our ability to manipulate information. Quantum computing raises the question of what information is at the deepest level, and what we are when we manipulate it. Classical technology extended human capabilities. Quantum technology extends them into a layer of physics that predates space and time as ordinarily understood. We have become, in a small and tentative way, participants in a computational process that may be the substrate of the universe itself.

And none of this, it must be said, is the deepest question. Not the encryption threat, not the medical promise, not the economic race, not the philosophy of identity, not even the strange fact that life itself turns out to be running on quantum physics. All of these are real concerns, real opportunities, real ambiguities, but they are not the deepest question. The deepest question is one that the mathematics does not answer, and perhaps cannot answer. And if the answer to it is yes, then every conclusion we have drawn in this entire 2-hour journey will need to be rethought from the beginning. The deepest question is about the direction of the door. When we built these machines and when we configure them to borrow computational resources from places we cannot see, is the arrangement a one-way reach, or does the traffic flow in both directions? We are going to end on this question because it is the question no one has answered, and because the machines are running right now, and the question is still unanswered, and the decisions being made today do not wait for philosophy.

Do you think the modern world is moving too fast to reflect on what it is building, or is this simply how every major technology has always arrived? Tell the community what you think below.

Chapter 14. Is the doorway a one-way mirror?

There is a silence in the mathematics. It sits right at the place where the equations describe what happens when a quantum computation finishes. The mathematics describes with extraordinary precision how amplitudes from every possible computational path combine through interference to produce a measured outcome in our branch of reality. It describes how destructive interference cancels wrong answers. It describes how constructive interference amplifies correct ones. It describes the probabilities with which each final state will be observed when the qubits are finally read out.

What the mathematics does not describe, and has never described, is what happens in the other branches after the computation finishes. It does not say whether those branches are disturbed by our act of calculation. It does not say whether interference propagates in the other direction. It does not say whether, when a qubit is placed into superposition and manipulated inside our refrigerated chamber, there is any reciprocal effect in the branches that contain the alternative outcomes. The equations simply stop at the measurement. They tell us what we observe. They are silent about everything else.

This is not, to be clear, a gap in the engineering. The machines work regardless of how this question is resolved. Google does not need to know whether its Willow processor is disturbing other branches of reality in order to report the correct answer to a random circuit sampling problem. IBM does not need to resolve the metaphysics in order to ship Nighthawk. IonQ does not need to settle the question in order to network its photonic interconnects. The practical business of quantum computing proceeds without reference to what happens on the other side of the measurement. The mathematics gives reliable predictions for our branch, and our branch is where the commercial returns live. Whether anything is happening anywhere else is, from an engineering perspective, not a question the firm has to answer to deliver value to its customers.

But the question is not, for that reason, uninteresting. It is one of the most interesting questions in all of modern physics, and a small community of theoretical physicists has been thinking carefully about it for decades. David Deutsch has written about it in his books and papers. Various interpreters of quantum mechanics working in the tradition of Everett have tried to formalize it. Max Tegmark has touched on it in discussions of what he calls the Level Three Multiverse. Researchers working in the foundations of quantum theory published papers about it in journals like Foundations of Physics and Studies in History and Philosophy of Modern Physics.

No one has resolved it. There is no experimental protocol that could answer it because the entire definition of decoherence says that the other branches are inaccessible from ours. We cannot peer across. We cannot send a signal. We cannot measure what is happening there because the moment we could measure it, it would no longer be a separate branch; it would be the same branch as ours. Epistemically, the question is enclosed.

And yet, the question persists because if the full Everettian picture is correct, then every time we perform a quantum computation, the picture must include observers in the other branches. If many worlds is true, then when your Google Willow processor completes a run, there are not only the qubits in our branch that now contain the answer. There are also qubits in the branches that contain the alternative answers, and presumably observers in those branches who are reading those alternative answers and becoming entangled with those alternative answers and integrating those alternative answers into their branch's ongoing history. The full picture is a branching tree of observer outcomes, each one proceeding independently, each one convinced that its outcome is the only one, each one unaware of the others.

Whether any process in our branch disturbs those parallel observer branches. Whether our use of the multiverse as a computational resource has consequences in the places we cannot see is a question that has no current answer.

There are several ways to frame the question that avoid the language of disturbance or consequence and simply ask whether the many-worlds picture makes sense at all in the context of quantum computation. Some researchers argue that the picture is purely ontological, that the branches are real but non-interacting in any meaningful sense, and that the question of reciprocal effect is therefore a non-quest. Others argue that the very fact that quantum computers can exploit the branches to compute means there must be some kind of cross-branch interaction, at least during the calculation itself, and that the nature of that interaction is a legitimate topic for theoretical inquiry. Others argue that the language of branches is itself misleading, that we should speak instead of correlations and amplitudes and unitary evolution, and that the question of disturbance disappears when stated in the proper vocabulary. Each of these views has serious physicists defending it. None has won a decisive argument.

What can be said with confidence is this: The machines are running right now, at this moment, in refrigerated chambers and vacuum traps and photonic arrays scattered across every major research university and every major technology company and every major state-funded laboratory on Earth. Qubits are being placed into superposition, are being manipulated through carefully orchestrated sequences of gates, and are being measured for their outputs. Answers are arriving. Calculations are completing. Every one of those calculations, according to the physics that makes it work, involves the cooperation of branches of reality that we cannot see.

The architecture of the multiverse is being used as a computational resource, and the scale of the use is growing. In the past 8 years, quantum computing has moved from a laboratory curiosity to a networked commercial infrastructure. Sycamore with 53 qubits was dismissed by some critics as an engineering stunt. Willow with 105 qubits and below-threshold error correction is considered the first genuine step toward fault tolerance. Nighthawk is a platform. Majorana 1 is a new kind of object. The IonQ systems are the beginning of distributed quantum computation. The roadmaps from every major player point towards systems with thousands and then tens of thousands and then hundreds of thousands of logical qubits within a decade, possibly sooner. At each step of that scaling, the computation becomes more tightly coupled to the informational substrate of reality. The branches being borrowed from become more numerous, and whatever we are doing when we do this becomes more significant. If the door is one-way, the scaling is irrelevant. If the door is two-way, the scaling is everything.

Nobody is stopping to answer the question. The engineering continues. The capital flows. The roadmaps extend. Nations race to reach the fault tolerance threshold first. Corporations race to be first to market with useful applications. Research teams race to publish the next breakthrough. At no point in this entire process is there a moment where the industry pauses to ask whether what it is doing has consequences that the equations cannot describe. The equations describe our branch. The equations describe what we measure. What happens outside of that, in the vast majority of the structure the equations imply exists, is left as an exercise for the philosophers. And the philosophers, so far, have not produced an answer.

This is where we end. Not with resolution, not with reassurance, not with a final summary that ties everything together. We end with the honest acknowledgement that the most powerful machines humanity has ever built operate by reaching into places we cannot see, using resources we cannot inventory, for purposes whose secondary effects we cannot measure. We end with the acknowledgement that this reaching is growing, that the door is widening, that the tools being developed are unprecedented, and that the question of whether the reach is reciprocal has no current answer.

What do you do with that? What does a civilization do when it discovers that its most powerful tool is a window into a structure it cannot observe, that it can extract from but cannot visit, that it can use but cannot know what it takes in return? These are not rhetorical questions. They are the actual situation we are in. The actual position humanity occupies at the current moment of 2026. Looking into a future that is being built very quickly, without pause, by people who are mostly doing good work, without having asked for the universe's permission.

John Archibald Wheeler, in his last years, used to say that the deepest questions in physics were the ones whose answers would produce in the listener the reaction, "Oh, of course! How could it have been otherwise?" He suspected that the ultimate answer to the question of what reality is, once we arrived at it, would feel obvious in retrospect. Perhaps he was right. Perhaps when this final question, the direction of the doorway, is eventually answered, the answer will feel inevitable. Perhaps the answer will be that the door is strictly one-way, and every concern we have raised in this chapter evaporates. Perhaps the answer will be that the door is two-way, and every use we have made of these machines will be seen in retrospect as the opening of a relationship with the rest of the multiverse whose implications we are only beginning to understand. Perhaps the answer will be something stranger than either of these possibilities, something that does not fit inside the categories we currently have available. We do not know.

What we do know is that the machines are running. We know that the answers they return are correct. We know that the correctness of the answers, according to our current best physics, requires the existence of the parallel branches from which the answers are assembled. We know that every year the machines grow more capable. That every year the number of qubits increases. That every year the calculations reach further into the informational substrate. And that every year humanity becomes more dependent on the results. We know the door is open. We know the door is widening, and the only honest thing to say is that we do not know what is on the other side. We only know that the answers keep coming back, and that so far, they have been correct. So far.

Thank you for joining us on tonight's long journey through the strange and beautiful physics of quantum computing and the many worlds we may or may not be surrounded by. If this investigation kept you with us, please leave a like and subscribe to the channel so that more of these deep nighttime explorations can be made possible. We love reading the conversations that form in the comments. What stayed with you most from tonight? Is there a topic you want us to investigate next? Every investigation we undertake is shaped by the voices in our community. So tell us where you want us to go from here. Until our next journey, rest well, sleep deeply, and remember that the universe is stranger and more beautiful than almost any story we tell about it.