This video explores the intriguing world of quantum computing, an area predicted to revolutionize our technological landscape. The discussion begins by breaking down fundamental concepts of quantum mechanics, which inform how quantum computers function. Renowned professor Seth Lloyd shares insights on how quantum phenomena, such as the double slit experiment, illustrate the core peculiarities that drive quantum computation, including the wave-particle duality and the probabilistic nature of quantum states.

As the conversation progresses, it delves into the potential power of quantum computers, focusing on the concept of quantum parallelism and its implications for computation. Through quantum superposition and entanglement, multiple computations could be performed simultaneously, surpassing the capabilities of classical computers. The video emphasizes the challenge of leveraging quantum interference to extract meaningful results from these computations, likening the process to orchestrating a complex symphony.

Main takeaways from the video:

Please remember to turn on the CC button to view the subtitles.

Key Vocabularies and Common Phrases:

1. quantum mechanics [ˈkwɒntəm mɪˈkænɪks] - (noun) - A fundamental theory in physics describing the properties of nature on an atomic scale. - Synonyms: (microscopic physics, quantum theory, wave mechanics)

quantum mechanics is a powerfully precise but deeply unintuitive description of the micro world.

2. entanglement [ɪnˈtæŋɡlmənt] - (noun) - A phenomenon where quantum particles become interconnected in such a way that the state of one particle instantly influences the state of another, regardless of distance. - Synonyms: (interconnection, linking, correlation)

Atomic clock, because it uses entanglement between different spins and pieces of the quantum clock in order to measure time to an accuracy that is millions of times better than people had ever measured it before

3. superposition [ˌsuːpəpəˈzɪʃən] - (noun) - A principle of quantum theory that describes a challenging concept — a physical system exists partly in one state and partly in another, a combination of all its possible states. - Synonyms: (overlay, combination, state mix)

Your single particle that you just made reference to in some quantum mechanical sense, passing through both of these openings at the same time.

4. wave-particle duality [weɪv ˈpɑːrtɪkl ˈdjuːəˌlæti] - (noun) - The concept in quantum mechanics that every particle or quanta can be partly described in terms not only of particles, but also of waves. - Synonyms: (dual nature, duality, particle-wave theory)

...many things about his intuition that we now know were misleading. I mean, he was brilliant. And it's a wonderful case study of how a deep thinker who's penetrated reality farther than any other human being can nevertheless have biases that take them in the wrong direction.

5. interference [ˌɪntɚˈfɪərəns] - (noun) - The phenomenon that occurs when two waves meet while traveling along the same medium. - Synonyms: (intervening, obstructing, disruption)

When the peaks add up and the peaks, you get positive interference

6. classical intuition [ˈklæsɪkəl ˈɪntuːˈɪʃən] - (noun) - Refers to the basic reasoning or understanding developed from everyday life experiences and conventional science, contrasting with quantum intuition. - Synonyms: (natural reasoning, common sense, standard thinking)

Well, you know, my classical intuition, and pretty much everybody's right, because, like, the intuitions where this stuff gets established when we're little tiny babies says, oh, okay, just the same kind of thing.

7. probability amplitude [ˌprɒbəˈbɪləti ˈæmplɪˌtjuːd] - (noun) - A complex number used in describing the behavior of systems in quantum mechanics, wherein the square of its modulus gives a probability density. - Synonyms: (probability magnitude, wave intensity, quantum amplitude)

So for those who like the technical language, it's a probability amplitude versus a probability.

8. quantum algorithm [ˈkwɒntəm ˈælɡərɪðəm] - (noun) - A step-by-step procedure or formula for solving a problem, capable of leveraging quantum mechanics principles. - Synonyms: (quantum solution, computational procedure, quantum program)

And if you are clever in building your quantum algorithm, that unique answer that we see being built up from all these quantum waves will be the answer that you're looking for.

9. linear algebra [ˈlɪniər ˈældʒɪbrə] - (noun) - A branch of mathematics concerning vector spaces and linear mappings between such spaces. - Synonyms: (vector math, matrix algebra, linear mathematics)

And this actually then becomes very potentially very useful because 2008, we didn't really have, like deep learning and stuff like that, and machine learning, let alone like, scary, but somewhat stupid, like large language models, but all the mathematics that goes, or great majority of the mathematics that goes into this processing information for machine learning and AI is actually phrased in terms of these linear algebraic problems.

10. quantum cosmology [ˈkwɒntəm kɒzˈmɒlədʒi] - (noun) - A field of study that applies quantum mechanics principles to the universe as a whole. - Synonyms: (cosmological quantum theory, quantum universe studies, space-time quantization)

The number of possibilities they can explore is more than the number of elementary particles in the universe observable. Universe observable. Thank you. Thank you. These, like quantum cosmology, people buy them.

Quantum Computing - Hype vs. Reality

Thank you. Thank you, and glad to see you all joining us for this discussion of quantum computing, an arena of technology that really has an enormous potential to transform how we live. Now, look, quantum mechanics is a powerfully precise but deeply unintuitive description of the micro world. And when you try to use ordinary language, which, of course, it evolved to describe the things that we do encounter in the everyday world, you find descriptions of quantum mechanics that involved statements like an object can be simultaneously here and there and everywhere, and you find statements that jump off from that. When you apply these ideas to quantum computers, you hear things like quantum computation creates every possible computation simultaneously. And you find things like quantum computation will result in an exponential increase in computing power, laying all of classical computation to waste. And you find statements like, all of this is right around the corner. Now, the fact is, there is some truth to all of these statements, but if you take them at face value without digging a little deeper, it can be quite misleading. And that's why I am thrilled today to have conversation with my friend and colleague Seth Lloyd, an expert in quantum computation, to help us sort this all out.

So, Seth Lloyd is a world renowned professor of quantum mechanical engineering at MIT and is currently working with a variety of groups to construct and operate quantum computers and quantum communication systems. Hi, Brian. Yeah, it's great to see you. Great to see you. All right. So what I thought we would do is first discuss the basics of quantum mechanics, just to sort of lay the groundwork for the entire subject. Because, look, as we know, the subject is so strange that even if you've heard it a hundred times and you don't actually work on it, it slips away. So I think that's a good thing for us to do to get things going. We can then move on to the basics of quantum computation, and then by the end of our conversation, which will be probably 6 hours from now, we can talk about what it really will take for this field to realize its full promise, something, of course, that you've thought a lot about. So if we can jump right in just to sort of motivate quantum computers, this quote of Richard Feynman that, you know, the one I'm referring to, if we want to understand the world, we have to understand in a quantum mechanical way, because by Jiminy and by golly, I think he said something stronger than, yeah, I think it was, you know, it was. Dammit.

Right. Dammit. There we go. Classical dammit. Exactly. If you want to, you know, make a simulation of nature, it should be a quantum mechanical simulation, because after all, that's how the world is actually based. And so that certainly gives some motivation for trying to put quantum mechanics within our computational architecture of how we do things. He also had another statement about a certain system that captures the heart of quantum mechanics. You know, the system that I am referring to, I believe that Feynman said everything about weird about quantum mechanics can be understood in terms of the double slit experience. The double slit experiment. Exactly right. So why don't we jump into that just as a sort of brief review? Because it really does capture the essential qualities of quantum mechanics, and that will then set us up to talk about quantum computers. So, if we look at this first animation here, tell us what we're seeing here.

So, this is a gun shooting classical particles through two slits. And the slits allow the particles to get through. And behind on the screen is a place where the particles land. And you can see the two images of the two slits made up there by where the particles have stuck on the screen. And this is exactly what our intuition would tell us should happen. Right. If what's coming out of the gun is particles, this is what should happen. And if it is large particles like pellets or bb's, that is what does happen. Absolutely. Now, we've gotten used to being able to extrapolate our understanding of the everyday world into realms that we don't directly inhabit. And were you to do that, you would anticipate what? So if you're now firing little particles, what would your intuition tell you? Well, you know, my classical intuition, and pretty much everybody's right, because, like, the intuitions where this stuff gets established when we're little tiny babies says, oh, okay, just the same kind of thing. It's like the particles are smaller, but the shadows and the light parts of the screen are the same.

But the amazing thing is, and this is really the heart of Feynman's observation of how this experiment captures the key weirdness. What happens if you actually do this experiment with little particles? Well, you find something really weird, which is that you don't just get the images of the slits, you actually get many more images. Moreover, actually, the lines where those images are or don't actually even lie in direct line between the gun and the slit, something entirely different is happening. Yeah. And this is completely unexpected, based upon classical intuition. And, in fact, I it was discovered, as I understand the history, actually, accidentally. Right. It was. I think it was at Bell Labs in New Jersey. Yeah. I believe that they were like, you know, they're blowing electrons off a cathode or, like a cathode or a tube, and it's like, hey, what's going on? Where are they, like, showing up on that screen? Yeah. And so this was in the 1920s, I think it was like 25, 26, 27 around that time when I. When the data began to suggest something like this. And kind of, you know, amazingly, at the same time, the theorists, motivated by other weird experimental results having to do with the spectra of atoms and things of that sort, had begun to introduce a new idea into the description of particles, which is this idea of waves.

The key quality of waves is that when two of them overlap, something intuitive happens, but something that is relevant to this unintuitive situation. So this is Central park. Throw a couple pebbles into that lake there. And what happens as those waves overlap? Well, because waves and water just add up where they are. So the peaks. When the peaks add up and the peaks, you get positive interference. When the peaks add up with the troughs, you get negative interference, and it all goes away. When the troughs are together, it becomes even more negative. So you get these interference patterns, which actually, any kid throwing a stone into the duck pond in Central park will see.

Yeah. And so that then suggests, in a way that is still deep and mysterious, that to explain that weird data that came from firing electrons or photons at the barrier with the two openings that we might need to go over to a description kind of like this, talk us through what we're seeing. Yeah. So here, this is what would happen if you're just sending waves through these slits. And then what you'd see is, because when wave goes, it's going through this slit and this slit simultaneously. If you extrapolate to where it lands on the screen, it may be that the wave coming through here lines up the peak lines up with the trough here, and nothing is there. But it could also be that the peak lines up with the peak there. In that case, the wave gets bigger. So what you see is what's called an interference pattern, and where the waves show up on the screen exhibits this interference pattern. That's not so weird for waves. That would happen if you just put two slits down in the dark pond in Central park. Exactly. So this blue wave.

And if we can just keep this up on the screen for a while, because I think this really requires a little bit of unpacking the waves that we see there, as you're saying, you could think of them as the water waves in Central park. But back in the 1920s, the suggestion was, yes, we're talking about waves, but we're going to now use them to describe particles, and you say, well, how do you do that? And that is what introduced this weird idea of probability. So how should we think about those waves in the quantum, quantum mechanical application? Yeah. So this notion that somehow particles have waves that are associated with them and actually, vice versa, things that are. Waves are actually made up of particles. Particles. Yeah. That's called wave particle duality, which already sounds like rather strange. And this is like, this is, in fact, the hundredth anniversary of the Schrodinger equation. So, like, coming up, I guess, 2026. Well, he actually was working on 2025, and he was actually doing the work work then on 2025. But Heisenberg would want you to really emphasize that he kind of got there first with another. But that's a separate issue. So, absolutely, we'll let them duke it out in quantum heaven. So, yeah, it's.

Yeah, so quantum weirdness in the form of wave particle duality has been around for a century or more, even before the Schrodinger equation, this idea that somehow waves were associated with particles, particles with waves in some weird way that nobody understood was already there. And Schrodinger's great discovery, which was also Heisenberg, contributed this in a different formulas. Schrodinger's discovery say, hey, there are waves out there. Waves obey some kind of wave equation. Like water waves, they propagate in a particular way. And he said, what's the equation that describes these waves? And this became, this is the Schrodinger equation that tells you how particles like electrons, how they actually follow these waves as the waves move around. And so when we look at those waves, if we then use the interpretation that I believe historically came from Max born, if we think about those waves as waves of likelihood, waves of probability. So where the wave is big, there's a high probability of the particle landing at that location, where the wave is small to low probability. So the interference pattern from central park is now being reinterpreted here, as it's very likely that the particle should land where the wave is big, which should give you the band structure, and where the trough and a peak have crossed and the waves canceled each other out, virtually no probability for the particle. And that then explains what's going on there.

But what does this do to our notion of reality? Oh, yeah. Reality and quantum mechanics and the human brain have always been at odds with each other, and it's not quite clear who's in the wrong. My suspicion is it's the human brain. I agree with you because, you know, our human intuitions are just not built for this kind of thing. Yeah, they're built evolutionarily to allow us to survive. And you don't need to know that quality of the micro world in order to get your next meal in the african savannah. Right. So that's why this isn't intuitive to us, but this appears to be the way the world actually does behave. Oh, yeah.

I mean, in fact, I think that many people, everybody finds quantum mechanics counterintuitive. Einstein famously found it counterintuitive. He got his Nobel Prize for quantum mechanics, for the photoelectric effect. Essentially says, oh, hey, these photons, they're waves, but they're actually made up of particles. But it was so counterintuitive that he never believed it. Because if you're Einstein, you get to trust your intuition. Right? And there are many things about his intuition that we now know were misleading. I mean, he was brilliant. And it's a wonderful case study of how a deep thinker who's penetrated reality farther than any other human being can nevertheless have biases that take them in the wrong direction. Right? Absolutely. I mean, I myself, my intuition is pretty much always wrong. So I don't care. I accept this. quantum mechanics, sure. Wave, particle reality, what the heck, you know? All right, but is that facetious, or do you really feel that? I mean, does this. I mean, we all get used to these strange ideas because we stare at them for 30, 40 years. But do you feel comfortable with this? Do you view it as an algorithm that you just sort of say, okay, it appears to work, or do you feel that that is the right description of reality?

Well, you know, if you go and live in a foreign country, many things seem weird and counterintuitive. But then if you learn the rules and both written and unwritten mechanisms, what goes on? And if you want to make something done, get something done in this foreign country, you have to set aside your own intuitions and do it how it is. So I and my colleagues are going to be hearing some of them later today have managed to kind of like, weird, but we know how it works, and we know the mathematics behind how it works. And then if we can trust our intuitions about the math and we do the experiments, then, by God, it works. So after, you know, a few decades of that, and by the way, I'm pretty sure that this is like the 16th anniversary of you and me talking about this at the World science Festival. Could well be. We talked about it the very first world science festival. Is that right? Yeah. Yeah. So, yeah, it's a long time. And so. But. And lots of progress has taken place, which you guys are going to hear about today because actually is really amazing, because now it's gone from just, like, you know, understanding what's going on at a theoretical level or building the very first simple quantum computers, like I did with my colleagues at MIT back in the 1990s, to building, you know, devices that aspire to be industrial quantum, industrial era devices. So a huge amount has happened.

So I'd love to get there in just a moment, but there's one more piece of background that I just want to extract from this example. When that probability wave hits that detector screen, it doesn't literally result in those bands appearing. Those bands are built up dot by Dot, because when the detector interacts with that wave, something happens with that interaction. Just take us through that, as that's going to be pretty vital in trying to apply these ideas. Yeah, absolutely. And on this picture right here, if you were to lower the intensity of the particle beam. So with this, it shows that the waves are quite intense, and this is how the waves add up there. But if you were to lower the intensity until there was even just like, one particle, the wave corresponding to one particle passing through both slits at once.

And we get into what you're saying about, like, quantum mechanics, stuff can be in two places at once, which you not to telegraph it, if you think of this as a bit, an electron over here is zero, an electron over there is one, then electron here and there at the same time is zero, and one at the same time. It's a quantum bit, or qubit. You go down to the single particle level. Then what you'd find is that the particle would show up somewhere on the screen, and you'd just get one speck where the particle showed up. But the spec would be in one of these, where these bright lines from the waves are. So, and then you build up many particles. You get this probability distribution that's shown on the screen.

All right, so just as a sort of quick summary of what we've extracted from this example, we have seen that particle dynamics needs to be described in a different way in terms of these waves of probability. We've seen, as you just emphasized, reality is embracing kind of many possibilities at the same time. Your single particle that you just made reference to in some quantum mechanical sense, passing through both of these openings at the same time. We've also noted that these probabilities are somewhat different from the ones at casinos, because, as you emphasize, they can kind of cancel each other out. They're not just positive numbers, right? Yeah. The waves have both positive amplitude and negative amplitude. That's like a negative probability in some sense. In some weird sense. That's right. So if you try to assign probabilities to all possible events, you find that. But actually, this goes under. Many people have also heard about the uncertainty principle. You can't measure position and momentum at the same time. And though if you try to assign probabilities to position and momentum at the same time, you'll find often these probabilities for events are negative. It's a funky thing. But the probabilities for events you measure, like where a particle shows up on the screen, is always positive.

Yeah. So for those who like the technical language, it's a probability amplitude versus a probability. It's a mathematical structure. But that's the essential idea, because, again, these quantum waves can interfere, and in that way, the probabilities can go up, they can go down, they can be turned into zero. And then finally we have this idea that when there's a measurement or an interaction, which is what that detector screen does, as that wave reaches it, you get back to your particle description, that range of possibilities is collapsed. And that is the essence of what's going on there. Now, if we now want to get closer to quantum computers, the subject that we're talking about, we want to instantiate those ideas. We want to realize them in a particularly evocative physical way. You already made reference to going through the left slit could be like a one in a computer. Going through the right could be a zero. So you sort of see that kind of binary digit emerging.

Another context in which we can do that is this idea of particle spin. And that's often how we talk about it, both in the language of quantum computing, but also, you know, in a quantum mechanics course, this is bread and butter of what we do and trying to look at very simple quantum systems. So if you could just sort of take us through that idea first, the idea of quantum spin, I mean, the word spin makes intuitive sense, but just give us the particle version of that. Should we think of it as a particle spinning around? Is that too coarse a way of thinking about it? You can think of quantum spin as a particle spinning around. So like an electron spinning around like that? Yeah, they typically call, if it's spinning like that, then you call this spin up. If it's spinning like that, you call that spin down. So the arrow is pointing in the direction my thumb would be. If the electron is spinning around like that, and the other one spins around like that on my computer. Now, these days, if I, like, go like this, and I do a lot of this in my meetings, then all kinds of, like, thumbs up and fireworks start going off and stuff like that.

It's not happening here. What's going on? Okay, there we go. Here's the particle going spin down. And so there, you can just think of this as like a. Like a basketball that's spinning around. Yeah. I'm sorry about the Knicks, by the way. The Celtics made it to the NBA Finals. The Knicks were a great team this year, by the way. I was watching that last one. I don't really follow, but my son does. Yeah. So it's like a basketball. And so just in terms of spin, you can think of it as spinning like this and or spinning like that, but unlike a basketball, kind of the same way you go down to this, like, single particle level with the waves, it's possible for a particle spin to be both spinning up and spinning down at the same time. You can't really visualize this. This is trying to visualize. It's flashing back and forth in a regular fashion, but that's not what's going on. It's both up and down simultaneously, very hard. Classically, on a classical. Yeah, yeah. We're like. I want to compliment the world science festival technicians for coming up with some great visuals to visualize something which is essentially unvisualizable. So, yeah, so it's spinning up and spinning down at the same time.

And if, as you say, you kind of force it to declare whether it's spin up or spin down, probabilistically, it will end up spin up or spin down. There we go. Great. But actually, before you actually let it interact with some measurement apparatus, these measurement apparatus, the original one, is called a stern Gerlach apparatus, where if the electron goes through a magnetic field, it will go up like that if it's spinning up, and it will go down like that if it's spinning down. And so if you make a measurement of where it shows up on a screen, you're correlating, or actually, again, telegraphing things, entangling position and spin. And then, so where the electron shows up on the screen tells you whether it's spin up or spin down. But until you make it interact with its environment like this, it's both spin up and spin down at the same time in some weird, funky, quantum mechanical way that I don't intuit. So we've seen examples where the particle can kind of go through both openings of the double slit, sort of at the same time, a particle can be both spinning one way and the opposite way at the same time again in some weird quantum mechanical way that is experimentally established. I want to get to one other, really. Could I just say, look, I mean, I know it's weird. You've worked on this field a lot. I've worked on this field a lot. It's very counterintuitive, but it's so well established that you just got to suck it up. I mean, that's just the way it is. Yeah.

And I think you really do need to chalk it up to the fact that we just don't need to know this stuff. And so it's not in our intuition. You know, I think, like, you made reference to these ideas are ingrained when you're very young, when you're a baby. You know, if you had a baby that was somehow born at, like, you know, ten to the -18 meters, you know, and they lived in that microscopic world, or, you know, it's probably child abuse, but if you had some VR realm where you made your kid grow up in a quantum mechanical environment, maybe it would be completely intuitive. Right? I mean, I can imagine that that's the way the brain is malleable enough to take in these ideas where you'd feel them, as opposed to just have to accept them. Yeah. In fact, I think that. I believe my daughter Emma is in the audience right here. And when she was a little baby, I have to confess that. You tried this. Well, I took her. No, I took her to being a good MIT professor. I lent her to my colleague Liz spelke at the MIT infant cognition lab. You know, like, because that's like, what does professor do with their colleagues? They exchange babies. And the way that this works is, I bet everybody here knows this about little kids. Or you may not know it consciously, but you remember it when I tell you. So, before the age of around three months, or three months is roughly when the age of object permanence is established. So the experiments that Emma was participating in were there was a little theater with a curtains, and behind it was a toy, and you'd close the curtains, the toy was there, and then you'd open it, and the toy might not be there now, before the age of three months, if you do this with a kid, close the curtain, there's a toy. Open data. Yeah. Yeah. I haven't been here for long. Guess that's just the way things are right now. It's this time in life that you were describing to. But after age of three months, you close the curtains, you open up.

I mean, I can't tell you. It's like it's obvious that they know something's wrong. Right, right. Yeah. Then by three months, your intuition for quantum mechanics is out the window. Right. Yeah. So you got to really have an early intervention if you want your kid to be quantum mechanical, you know? So I now want to get in to applying these ideas, which, from a 30,000 foot level, is kind of straightforward to get the basic idea of quantum computing. So, you know, if you have these kinds of blended states of a zero and a one simultaneously, then you can do computation, presumably in a very different way. So, just to sort of give us the. The first example, just tell us what we're seeing here and how to distinguish the left column and the right column.

Yeah. So, classically, classical computers work by operating on bits, and a bit is the abstractly, it's the distinction between two possibilities, heads or tails, true or false. But traditionally in computer science, called zero and one, it doesn't matter what you call them, it's what you do with them. And the word bit also refers to the. The object in a computer. In our cell phones and other supercomputers, they're little transistors that are, like, switched this way or switch that way. And we've already seen when we go to the quantum mechanical realm, starting with the double slit experiment. Electron going through the right hand slit is zero. Electron going through the left hand slit is one. But these electrons just very naturally go through both at once. And so that is a bit. That is both zero and one in some funny quantum way that we don't intuit, have an intuitive grasp on. So we call that a quantum bit or a qubit. And similarly, I mean, actually, when I first started making models for building quantum computers back in around 1990, I realized, hey, any possible quantum degree of freedom is like this. So spin is like that as well. Call spin up zero. Call spin one down, one spin up and down at the same time as zero and one at the same time in some counterintuitive quantum mechanical sense. So then the question arises, and this was really, by the way, first raised by David Deutsch around 1985, that suppose we actually have a computer, a digital computer, but it's operating on qubits instead of bits. What might it be able to do?

And if you think about those, say, as operations, can I think about the one on the left, as that classical computer can do this operation or that operation. But in some sense, the quantum computer is doing both of these operations. Is that a reasonable way of thinking about it? That was well said. Yeah. So if you think about how does a classical digital computer work? Well, so it takes all this information and busts it up into bits. So, like a photograph or a movie, there are all these pixels. Each pixel gets, like, 32 bits of information about what's going on in that pixel. And then to process that information visually, like with one of these lovely new AI visual programs where you can use your brain to actually make the computer do make a picture of Brian, who looks like a cat or something like that, then it just busts up these bits into the information, these smallest bits, and then flips them rapidly in a systematic fashion. And so that's all that's really going on in a digital computer. And then in a quantum computer, it's the same thing, but with quantum bits. So you store the quantum bits on electrons spin up and spin down at the same time. Photons like particles here and there at the same time. And then if you think about a computer, this bit, a bit could be an instruction. Zero could tell your quantum computer add two plus two. One could tell your quantum computer add three plus one.

But if you put a qubit that's zero and one at the same time into your quantum computer, it's going to be, in some weird quantum sense, adding two plus two and adding three plus one at the same time in a very counter intuitive fashion. And so that begins to speak of the power of doing many things at once. Now, with a single bit or qubit, it doesn't really feel like you're getting a huge amount of mileage of that. But as you start to scale this up, so if you had two qubits, right, you got one operation in the classical, but you sort of got four, you know, possibilities happening in the quantum. And you obviously see what happens here as the number of qubits starts to get larger and larger. You're starting to see one versus many, where many is getting increasingly large. I mean, just for illustrative purposes, if we put, like, 20 in there, you can't even list them all on the page. And so if we make sort of a little summary table that compares just the numbers on the classical versus the quantum side, I mean, that's pretty amazing. What's happening in that column on the right there. Yeah, it grows very rapidly and exponentially is the word. And so, actually, you know, you go two, 4816, 32, 64, and you get up very large rapidly.

So, in fact, a good. If you had a quantum computer that had 300 qubits where you could actually perform quantum computations, by the way, that's what we're going to eventually. That's where we're sitting right now with the actual quantum computers people are building, then the number of possibilities they can explore is more than the number of elementary particles in the universe observable. Universe observable. Thank you. Thank you. These, like quantum cosmology, people buy them. And so this, I think, is the origin of the statement that we hear kicked around a lot, that quantum computers will increase all calculations by this enormous exponential factor. But I want to, in our remaining time, dig down on that a little bit more, because, as you know well, and as I've learned from your community quite well, that can be a misleading statement. Right? Because, you know, if you're trying to undertake a particular task and you are following every possible computation, that might lead you to the answer you're looking at, well, how do you pick out the one answer from the many that all these many parallel computations will yield?

Yeah, I mean, that's right. And indeed, actually, when David Deutsch first proposed, like, using quantum computers to do what's called this quantum parallelism, that's the buzzword that goes along with it seemed, hey, we could solve really hard problems. We could search for all possible solutions for the way to get rich using cryptocurrency. Actually, people do propose using quantum computers to search for bitcoin. But the problem with that is what we were talking about before is, at a certain point, you have to get, you leave the quantum computer alone. You let these qubits do what they want. You don't want to look at them, because when you measure them, they have to declare mi zero or mi one, and you lose their power. At that point. You lose their power. That's right. In fact, it's considered very rude to look at somebody else's quantum computer in the laboratory while it's operating, because if you sneeze, the quantum computer gets a cold. But, yeah. So all you do is just, you set all these computations in parallel, quantum parallel, and then you just make a measurement, say, hey, you know, what's the answer? You'll just get some answer at random, completely random. That's not very useful.

And so the goal is presumably to make use of this weird kind of probability that's inherent to quantum mechanics, where, as we said before, it's not just positive. You can have these interesting interference effects that if you were really clever, you'd be able to take the various waves and their various probabilities and somehow combine them so that the combined wave doesn't have all possibilities at random, but rather is getting closer and closer and closer to some unique answer. And if you are clever in building your quantum algorithm, that unique answer that we see being built up from all these quantum waves will be the answer that you're looking for, even though you did not know what the answer was. That sounds incredibly difficult. That was beautifully said. And when you said, I said, no, that can't possibly be true. Yeah, right. Yeah. It's. You have to remember, just even in the two slit experiment, the interference is actually very important in the two slit experiment and in the quantum computer, if you've got 300 qubits, you have a gajillion waves, and they're all wiggling up and down, and you want to get them to interfere with each other in just the right way. So it's not just doing, having all these waves at once.

You have to construct a kind of a harmony. A classical computation is sort of like a gregorian chant, like, whereas a quantum computation, you get this, hey, sing a note. Just go like, what do you want me to do? Just so you know, just go. Right. That was fun. Nice, right? Hey, okay, we got a widow. So we interfered. But in a constructive way, then. Yeah, exactly. So the sound of the chord. It was a very nice chord, by the way, the sound of the chord, it has a particular feel and quality of the sound that you instantly recognize. But that is not a property of one of these sounds, one of these waves on its own, it's the property of the interference between the two of them. So if, like, classical computation is like a gregorian chant doing only one thing at once, a quantum computation is a symphony, but a symphony more complex than anything Wagner ever wrote.

And so the goal then would be to take that symphony and have those various waves interfere. As you said, in the double slit, they're interfering in a very particular way, picking out various locations where the particle could land. But just sort of visually, in the quantum computer, if you had all these possible outcomes of your many qubits, you'd want to somehow choose those waves to interfere and begin to suppress most of the answers that you don't care about, because they're not the answer of interest and bring to the fore the few or maybe the one answer that you were actually looking for. Nice graphic like that. Yeah, exactly. That was well put. Exactly. Yeah. You've got to make, you know, you have to make all these different possibilities, but there are possibilities that exist simultaneously in a quantum mechanical fashion. You've got to make them add up to give you the answer to your problem, and that's tough to do. And how many quantum algorithms have, you know for important problems have been developed that can do, that, can pick out the answer, not knowing which one, but somehow combining the waves in just the right way, that the answer of interest is all that emerges at the end of the quantum computation.

Yeah. Well, when we first talked about this in 2008, I would say not so many. There's. There's famously, Peter Shor in 1994 showed that you could use quantum computers to factor large numbers, break public key cryptography, thereby striking fear into the hearts of agencies with three letters in their name, like NSA. Not that they have hearts, right? But. And then the other primary algorithm from that time, there are two others. One is searching. You could. Grover's algorithm. Grover's algorithm, which allows you to use interference of waves. Actually, Grover was inspired by phased array radar, where the waves are, like, adding up. You want to detect an object over here. The waves are coming out, you interfere them together to figure out where the object is. It's called phased radar. The quantum version of that is Grover's algorithm. That gives you an enhancement, quite a significant enhancement, in how accurately you can detect and image things. And then the third was an algorithm that I developed, inspired by Richard Feynman's original quote is like, yeah, world's quantum mechanical. God damn it. We need quantum mechanical devices to understand it, which showed how you could take quantum computers and use them to simulate other quantum systems, including laws of elementary particles, even string theory.

And so do you have an intuition about what kind of problems submit to a quantum algorithm that would allow the leverage of a quantum computer to reach its potential? And what kind of problems, we're always going to be stuck with a yemenite classical, one by one by one computation? Yeah, that's a great question. Of course, nobody knows the answer. We're all looking for new algorithms all the time. In 2008, my colleagues Aram Harrow, Avenaden Hastadim, and I realized that you could take this kind of like, the mathematics of describing waves and solutions to wave equations is called vector algebra. It's like you have these big arrays of numbers called vectors. You have big square arrays of numbers called matrices, and you can. The rules for multiplying them, and the Schrodinger equation and Heisenberg equation are phrased in terms of these laws of algebra. And we realized that you could actually use quantum computers because of this connection between the physical waves that are out there. These quantum mechanical waves and mathematics, you could use it to solve all these problems involving linear algebra.

And this actually then becomes very potentially very useful because 2008, we didn't really have, like deep learning and stuff like that, and machine learning, let alone like, scary, but somewhat stupid, like large language models, but all the mathematics that goes, or great majority of the mathematics that goes into this processing information for machine learning and AI is actually phrased in terms of these linear algebraic problems. So now we actually have a whole host of possible applications where if we could build a quantum computer, even a modest size quantum computer with like a few thousand qubits to rub together, then we could actually solve a bunch of various societally useful problems. And so when you hear a rough summary that we do here, that quantum computers will lay classical computation to waste, everything will be subsumed within a quantum approach. How accurate or misleading do you judge that kind of statement? Yeah, I don't think that people are going to be running Microsoft Word on their quantum computers anytime soon, actually. I don't use Microsoft Word anyway. But even if that'd be kind of a waste for quantum computers, actually, it's not clear for these hard optimization kind of problems, because you can't search through all these possible solutions at once. You need to restrict your attention to solutions where you can combine the waves using this positive interference that turns out to be a very large class of problems for things like quantum machine learning. Quantum AI, though I must say, I'm not sure I want to get into a self driving car that was powered by a quantum computer at this moment. And so there are potentially a lot of applications for it, but it's not going to change everything.

I think that if quantum computation does succeed, and by the way, it's not clear it's going to. Right. There's like, problems. I think we would discuss the problems of noise and errors and the sensitivity of quantum systems. We have to be able to overcome those. We're kind of a push comes to shove time where it looks like 50 50, we're going to be able to make these large quantum computers may happen or it may not. If it does work out, then it's going to be really useful, and I would say a pretty positive way. So, one final question, which I think really is really just intuition more than anything else. If you were to say a timescale for realizing the full potential of quantum computers, do you put that five years, two years, 100 years, or is it just too hard to say at this point?

Well, my prediction, I try not to make technological predictions because you particularly on camera, because you always look dumb, like few years down the line. But I think people are working very hard, lots of well known companies, governments are working very hard to do this because of the obvious benefits of doing this. It's a tough road to hoe and very uncertain. I say ten years plus or minus never. Well, okay, ten years, minus five, plus never. Yeah. Yeah. So you do hold the possibility that we may never actually realize in the manner that, at least theoretically, we can imagine. Well, so to be fair to quantum computing and quantum information. Quantum information. And quantum information processing is a much broader field than just building a quantum computer. So you can, like, play quantum video games, but because actually, if you look at problems like sensing and imaging and precision measurement, and then you're taking these quantum states by probing the world with light going on it, the quantum states come out, you process it. Quantum information processing allows much, much more precise ways of doing this. Already, Dave Weinland's optical frequency quantum clock, for which he got the Nobel Prize, is called the quantum logic quantum clock. Atomic clock, because it uses entanglement between different spins and pieces of the quantum clock in order to measure time to an accuracy that is millions of times better than people had ever measured it before. And same for sensing and imaging. Maybe it's possible to make something like a quantum magnetometer that might actually be able to measure the magnetic fields in the brain much, much more accurately so that you could actually gain a picture, a movie, a four dimensional movie. What's going on there? My colleague Michelle Riley is in the audience here right now, and we're working on this right now. And those applications are actually already here and those are here to stay. But, yeah, like striking fear into the heart of the NSA by factoring large numbers that may or may not happen. Well, it's been a fascinating conversation on the basics of the subject.

Thank you so much. Great. Thank you very much, Brian. Great to see you. Thank you.

Quantum Computing, Technology, Science, Quantum Mechanics, Seth Lloyd, Innovation, World Science Festival