Vlatko Vedral is Professor in the Department of Physics at the University of Oxford and Centre for Quantum Technologies (CQT) at the National University of Singapore. He is known for his research on the theory of Entanglement and Quantum Information Theory.
Steve and Vlatko discuss:
- History of quantum information theory, entanglement, and quantum computing
- Recent lab experiments that create superposition states of macroscopic objects, including a living creature (tardigrade)
- Whether quantum mechanics implies the existence of many worlds: are you in a superposition state right now?
- Present status and future of quantum computing
Entanglement Between Superconducting Qubits and a Tardigrade
Macroscopic Superposition States: entanglement of a macroscopic living organism (tardigrade) with a superconducting qubit
Please send any questions or suggestions to firstname.lastname@example.org or to Steve on Twitter @hsu_steve.
What is Manifold?
Steve Hsu is Professor of Theoretical Physics and Computational Mathematics, Science, and Engineering at Michigan State University. Join him for wide-ranging conversations with leading writers, scientists, technologists, academics, entrepreneurs, investors, and more.
Steve Hsu: Welcome to Manifold. My guest today is Vlatko Vedral. He's a professor of theoretical physics, both at Oxford University and the National University of Singapore. Recently, he and his collaborators, most of whom I think were experimental physicists, published an amazing result in which they created a macroscopic superposition state in which they created a superposition state involving a living being, a kind of tiny little bug called the tardigrade.
And we will discuss that experiment in some detail, and also talk about the implications of this experimental accomplishment for things like the foundations of quantum mechanics, quantum information, and quantum computing.
So welcome to the podcast, Vlatko.
Vlatko Vedral: Thank you very much, Steve. It's a great pleasure to be talking to you.
Steve Hsu: Yes, I'm really glad I could get you. And it's just amazing because I've been waiting, anticipating this kind of experimental result for many, many years, and I'm glad you guys finally accomplished it. And now through the magic of the internet and podcasting, I can just interview you and share the conversation with, you know, perhaps thousands of listeners.
Vlatko Vedral: I'm very happy. And I think it is a very exciting result for me, to speak about. You're right that it's an old idea in some sense, and it's taken a long time to realize, and also I'm excited about the future, you know, how we can extrapolate this experiment and what else we can do with it.
Steve Hsu: Great. We're going to get into all this stuff. And let me warn the audience that, you know, I'm a theorist and you're also a theorist, although probably know by now, you know, a lot about the details of the experiment, but of course in physics, the real work is done by experimentalists who have to really do things in the, in the lab and make things happen.
So for theorists to conjecture something or imagine something, but those guys have to do the heavy lifting to actually make it happen.
Vlatko Vedral: Yes, Actually all the credit goes to my colleagues in Singapore, in fact, who managed to achieve these amazing results.
Steve Hsu: Great. We're going to get into all of that, but let's start with your background because I think the listeners always enjoy hearing about the intellectual odyssey that each scientist goes through in their careers. So you are originally from, am I correct, Croatia?
Vlatko Vedral: Serbia actually from Belgrade
Steve Hsu: Oops, huge mistake. Wasn't it?
Vlatko Vedral: Well, not a huge mistake um geographically. I think we are talking about a couple of hundred kilometers away from Croatia. And so Belgrade, the capital of Serbia. And I went to school there as well. So I was in Serbia until I was about 19, I think, 18, 19.
And I think what influenced me was probably my high school intellectually speaking, that was the greatest influence. And it's, I think around the age of 15, 16, when I really started to get into physics.
I think it was maybe at that time for the first time taught as it were, and at a, at a level that could actually be exciting for a teenager. And I think that grabbed me. Around about that time I thought this was really mind-blowing and maybe that was the beginning of all of it for me.
Steve Hsu: Yeah, I would say the same thing. I think it was during my teen years where I realized just how amazing the concepts in physics were, and I just couldn't get enough of it actually.
Vlatko Vedral: Yes. Yes. I think a lot of us physicists feel very similarly about it, that there is a feeling that's very different. I mean, it's difficult to describe in a way why we feel this way. but even, you know, even the simplest concepts in physics, it's just mind-blowing that you can, you can actually make predictions about all sorts of things that at first sight look complex. But somehow you can write a few lines on a piece of paper, calculate things, and actually, then you can do an experiment and really confirm that. I found that mind-blowing. It's magical. I still find it like that.
Steve Hsu: Yeah, I agree. It's amazing.
Now, I believe both your undergraduate degree and your Ph.D. are from Imperial College in London. Is that correct?
Vlatko Vedral: Yes, it's correct. It was, it was kind of almost an accident actually that a cousin of mine, who, I think lived in London at the time, actually, I think since maybe the seventies, actually, I think she moved from, from Serbia to England when she was young as well.
She actually was married to a person who himself had a Ph.D. from Imperial and it's just by talking to him. I was talking to him about my passion for physics at that time. I think he was a biochemist. And he was a Ph.D. from Imperial himself. And actually, he said to me, if you like physics so much, why don't you come and study at Imperial? And he told me about all of the great researchers that he was aware of and the fact that they had, I think, five Nobel laureates in physics and all of these things. And he made it really sound very exciting for me, which is actually what determined my choice. That's how I applied to come to Imperial.
And then as you said, it was natural after my undergraduate degree to actually stay on and do a Ph.D. I had a brief stint at Oxford in between, and I nearly actually went to do my Ph.D. at Oxford. I had an offer from Oxford, but somehow it seemed to me better to continue at Imperial at that time.
Steve Hsu: Now, this was the nineties. Is that correct?
Vlatko Vedral: This was the nineties. Yeah. All in the nineties.
Steve Hsu: I see. So that, in my field, the, maybe the best-known guy at Imperial, I think if I'm not mistaken as Tom Kibble. Are you familiar?
Vlatko Vedral: fantastic guy. Fantastic guy. Yeah.
I think he was inspirational actually. and you're right. I think he comes from that. I think that's a group, the group of physicists who actually made Imperial famous. People like of course Abdus-Salam. I think he may have even started the theory department at Imperial.
Then Chris Isham I think who one of Salam's students, Tom Kibble, certainly that generation of people. And I think they really achieved their kind of a world-renowned research at that time. and it certainly was impressive to be, I think even though you were doing, of course, your undergraduate degree, and you're not directly involved with research, I think you are still fully aware of what's going on around.
And I think Imperial has, has certainly been an impressive place.
Steve Hsu: Yeah. So in the nineties, I'm guessing that, so quantum information wasn't already a big thing. So perhaps were you studying quantum optics for your Ph.D.?
Vlatko Vedral: That's a, that's a great question, actually, because it was you're you're right. I was somehow extremely lucky that in my last undergraduate year, what happened is that, I think this was 94, so Shor, Peter Shor published his famous algorithm. Now Shor's algorithm for factorization of numbers on a quantum computer.
So this was really the first piece of evidence that a quantum computer could actually do something exponentially more efficiently than any classical computer. This influenced me a great deal in my last year, because in your last year, of course, you kind of tend to get a bit more into research. You do an extended project. And my project was really on entanglement between photons and atoms. So that's something that I actually started with and it's kind of kept me intrigued over a long period of time.
I was lucky at that time. And it was quantum optics. You are right. And my, my, my supervisor ultimately was Peter Knight actually, who, who, is a quantum optician. But I was very lucky to meet through him, Artur Ekert who actually shortly before meeting him, he had developed entanglement based quantum cryptography. And through him, I met David Deutsch. And at that time that was more or less the kind of quantum computing community in, in, in the UK. That's how small it was.
And so I was very lucky at that time that the field had just started to kind of become extremely interesting. And of course there was this huge promise of exponential efficiency.
Steve Hsu: No, no. Those guys that you mentioned, are they not Oxford people?
Vlatko Vedral: Oxford people. Yes, I think, I think Imperial and Oxford have had a, kind of a long standing collaboration in, in many directions. And I think Artur Ekert was formally at Oxford, but I think he would spend almost like one day a week at Imperial as well. So somehow we felt like, like a very, very small family at that time.
I, you know, when I met David Deutsch, came to Oxford. I think I was just finishing my undergraduate degree and I think David gave me three or four papers to read on quantum computing. And it was interesting. It took me a bit of time to read them. I came back to him and I said, David, I read the papers, what do I do now? And he said to me, now you're an international expert on quantum computing. And it was very funny, you know, to a 22, 23-year-old myself to hear. That's how small the whole field was.
And I found it immediately extremely exciting. All sorts of aspects, both physics and the computational side as well.
Steve Hsu: That's an amazing story. I’m older than you, so I missed this boat. I remember when Shor's result came out, I was just transitioning from postdoc at Harvard to assistant professorship at Yale. And at the time I was spending a fair amount of time talking physics with a guy called Eddie Farhi at MIT.
Vlatko Vedral: Yeah.
Steve Hsu: And I remember discussing Shor's algorithm with him and actually trying to come up with other interesting things to do. And I think Seth Lloyd wrote his paper on quantum simulation right around that time.
Vlatko Vedral: Yes. That was a big thing as well. You almost on a weekly basis, you had a new proposal for how to implement qubits and gates, and it was extremely exciting. It was clear that experimentalists were starting to get into it and take it more seriously. It felt already at that time it felt like a field that was definitely going to explode.
Steve Hsu: Yeah, I think I missed the boat because I sort of stayed more focused in kind of particle physics and cosmology.
Vlatko Vedral: But still very exciting.
Steve Hsu: Yeah, I'll tell you a funny story. When I was an undergrad at Caltech, Feynman was teaching a class called limits to computation and you know, which was secretly a kind of a course about quantum computing. Well, before its time. And I remember saying to my dad, he, my dad asked me, well, what, what's Feynman doing, why aren't you taking a class from Feynman? And I said, well, he's teaching this weird stuff. He's not teaching, you know... I would have been very happy to take a course with him on path integrals or, or particle physics.
And I said, well, he's doing this weird stuff. It doesn't even seem to be physics. And that's how stupid I was as a kid.
Vlatko Vedral: Yes. It's very unusual. I think that's the course you attended probably came out as a book later.
Steve Hsu: Yes.
Vlatko Vedral: A beautiful book by Feynman.
Steve Hsu: Yes
Vlatko Vedral: Which sounds, as you said, like a weird mixture of kind of solid-state physics and computing logic as well. It's very unusual.
Steve Hsu: Yeah, but he was thinking through all these things at the time.
Vlatko Vedral: Yes.
Steve Hsu: We now really appreciate the fundamental nature of these questions. Deutsch was also very early in understanding these things.
Vlatko Vedral: Yes. I think two of them probably made the biggest contribution. You're right. I think the idea comes from Feynman and then somehow this idea that you can actually make a universal quantum computer and you can kind of offer this blueprint through gates and qubits, comes from Deutsch. And I think it's why the whole thing took off ultimately.
Steve Hsu: Yes. When I was in this phase when I first moved to Yale, I remember going down to the library and I found some, I forgot what the journal is. It's some British journal where his papers on the quantum Church-Turing thesis. I remember reading these things and it was very mind-blowing to me because it was totally different from the kind of focus of most, you know, particle physicists or quantum field theorists, just totally different.
Vlatko Vedral: Yes, I think it is very different. And traditionally, as a physicist, you don't really learn these things in your undergraduate degree. You know, you don't talk about logic. You don't talk about information theory, almost the tool. I was lucky to, again, going back to my high school, you know, I was lucky that I had some kind of rudimentary understanding of logic, mathematical logic.
So I did a little bit of that as part of my own maths and, maybe that's what made it a little bit less unusual when I was reading these papers, but I think otherwise you are right. It's um it's very, very kind of unorthodox for a physicist.
Steve Hsu: Yeah. You know, in high school I read this book called Godel Escher Bach.
And so I was familiar with Turing's work and Gödel's theorem. So I was quite interested in that stuff, but I had sort of made a conscious choice to do physics instead of that stuff.
So kind of mind-boggling that quantum mechanics could possibly be connected to these concepts.
Vlatko Vedral: Yes.
Steve Hsu: I just want to say one other thing, which, and get your reaction to this. So there's another thread here, which is that a bunch of people who come more from particle theory, string theory, gravity who spend a lot of time thinking about the quantum mechanics of black holes.
So whether information is destroyed by black holes and, you know, the evaporation of black holes destroys unitarity. And so from the group of people that I know well, a lot of the awareness concepts in quantum information theory and things like that came from believe it or not from being interested in the physics of black holes.
Vlatko Vedral: Yes. Yes.
Steve Hsu: I don't know if that's apparent to people on the other side, like to a guy who does quantum optics, you know, somebody coming to them saying, oh, I got interested in your subject because of black holes. It just sounds totally crazy.
Vlatko Vedral: No, no, I think not at all actually, because I think even certainly myself at that time, but I would say even my colleagues around me, we were definitely aware of [Jacob] Bekenstein's work, right, on the information. And I think that you know, this area where you have thermodynamics, information theory, quantum physics, I think a very rich area. and certainly most of us were even at that time aware of lots of this work in that direction. And I think it's a natural connection between thermodynamics, I would say and information theory. It's fascinating work.
And then you mentioned Seth Lloyd, of course, he's had these speculations and in fact, doing a simple calculation on, on thinking of a black hole as, as the kind of ultimate computer, you know, and how much, how many quantum bits can you store in, in a black hole and how quickly can you actually execute the gates?
I think those are fascinating questions, but more fascinating probably is what you brought up, which is actually the physics side of things. You know? It's a place where quantum physics and gravity kind of meet equally, strongly. And then of course there is this fundamental debate, you know, which one wins and what do we do in this domain?
and will there be, I guess, string theory center there advocating that there will be maybe a theory that supersedes both and, and, you know, gravity and quantum mechanics become somehow special cases of this possibly. So it's still a very exciting direction.
Steve Hsu: Yes, actually, I hope at the end of our conversation to come back to maybe some of these ideas. But maybe we should just get into the experiment so the audience can get a better feel for it.
So maybe in your own words, just tell us maybe the history of this experiment. Like, how did you get involved? What were the motivations for it? And then ultimately what was accomplished?
Vlatko Vedral: yes, I think they, like I mentioned, the idea really is I would say like most of these quantum experiments, they really go back to this thought experiment of Schrödinger's. Yes. And in fact, one could even say that all quantum experiments are really more or less complicated versions of Schrödinger's idea.
So Schrödinger had this idea, that if you have an object in a quantum superposition, you know, if you have an atom that's either decayed, emitted a photon, or not emitted a photon and quantum [unclear] exists at the same time. He then thought that anything else that couples to this entity would have to somehow join into this superposition. So of course, as we know, Schrödinger had this bottle of poison that if the atom decays breaks the bottle, and then there was the famous cat that gets poisoned. But in the other branch of the superposition, the atom has not decayed and nothing happens to the poison. It doesn't get released and the cat is perfectly happy and alive and all is fine.
So the point that Schrödinger was making is if you take quantum mechanics seriously. And I think at that time, you know, this was 1935, the way he wrote this is almost in a, in a way that he was exposing an apparent paradox. So I think if you read that paper in 1935, you would think that Schrödinger probably thought that this would not go like this.
However, I think even he changed his mind, and I think towards the end of his life, in the fifties, if you read his writing at that time, it seems to me, he perfectly-acknowledged that this kind of entangled state would be possible in principle. And in fact, I think he advocated this as the most consistent view of quantum mechanics.
So the idea is always, if you have a simple object in a superposition, and if this object somehow interacts with another object, then that other object also has to become part of that superposition. Unless of course, something happens that's different to quantum mechanics and that would prevent this superposition.
So what Schrödinger was actually offering is also a test of quantum mechanics in a way.
My own interests. So that's kind of the experiment that I think captures a lot of these things. And like I said, Probably at the time and Schrödinger was talking about this, this was probably considered science fiction more than a realistic possibility.
But I think with time as the technology really improved for me, I think the transition was, at the end of the nineties. I think throughout the nineties, most of the experiments that were done in the quantum optics labs where maybe entangling two photons, maybe the internal degrees of freedom of photons, maybe part maybe frequency, you could entangle a couple of atoms. Maybe you'd be entangling photons with atoms, but it was always at the level of a couple of particles and not more, that was the kind of the state of the art at the beginning of the field.
However, around about the year 2000, just at the beginning of the new millennium, people in the solid state, in the many body physics direction started to take computing and quantum information more seriously. And I think this was a game changer. This really was because I think most of us now started to think, well, maybe, this quantum entanglement phenomenon could actually manifest itself in the microscopic domain. And in fact, as you know, the leading current technologies for qubits are really superconducting qubits. They are in the macroscopic domain and they were crucial for the experiment that we are discussing here. So I think that started to develop around about the year 2000 onwards.
In fact, it's an interesting anecdote that I think many people before that time, would even be betting that you couldn't make a qubit out of a superconductor that you couldn't control all the relevant, superpositions, appropriately to make it into a full fledged qubit. And then I think 1999-2000 people started to do exactly that. And I think that's, that was crucial. I would say for this kind of experiment.
Steve Hsu: So just to back up a little for the audience and I think you're, you're, you're, you're describing things at just the right level, but there are many listeners to the podcast who aren't actually physicists. So
Vlatko Vedral: Yes.
Steve Hsu: I'm going to back up a little bit and just explain things to them. So a unique aspect of quantum mechanics is this idea of an entangled state or a superposition state, where to take the simplest case, maybe you have the spin of an electron. Maybe it can have a spin pointed in the up direction and spin directed in the down direction. And somehow in quantum mechanics, you can make a state where it's in a sense simultaneously in both the up and down state in a very specific way.
And the period that Vlatko was talking about, the early nineties, late nineties, we were already pretty experienced with making superposition states out of really simple elementary degrees of freedom, like the spin of an electron or the photon polarization. But it was still then an open question, whether you could take a macroscopic system, a system with many degrees of freedom place it in a superposition state.
And I think Vlatko just told us that already the qubit, some of the most promising qubits that might be used in the future for quantum computers themselves macroscopic, superposition state. So these superconducting qubits involve, correct me if I'm wrong, Vlatko. They involve, for example, the circulation of charges in some superconductors. And so you could have a superposition between charges that are circulating clockwise and charges are circulating counterclockwise in some kind of a superconducting system.
Vlatko Vedral: Yes. I think that was the key difference. Exactly as you described it, that prior to that you would maybe entangle two spins of two different electrons or two different nuclei of atoms, or you would entangle two photons uh, when, when you know, polarization would now be the relevant degree of freedom, it's kind of like the spin of, of a photon. but no one did anything more impressive than that.
With superconductors, of course we are talking, one qubit is the one you described where you have a whole current moving in two different directions, clockwise and counterclockwise simultaneously. Or you could think about two tiny islands of this superconductor, and you could think about different numbers of electrons existing on these two islands, which is also kind of like a more macroscopic property than anything before that.
And I think this was crucial because like I said, there were actually skeptics before that thesis who would think that nothing like this could even be achieved. And I think this technology, you know, proved to be crucial at that time.
Steve Hsu: So those systems that you just described, the islands or the rotating charges. Th those are, they involve many degrees of freedom. Say for example, millions of degrees of freedom, but they're not yet living creatures.
Vlatko Vedral: That's right?
Steve Hsu: And, you know, part of Schrodinger's original thought experiment was that the ultimate thing that gets put into the superposition state is a cat, or maybe even the experimentalists who look in at the cat. So that's going to be the ultimate question. But already there was resistance by people saying that, okay, if it's got a million degrees of freedom, you will not be able to put it in a superposition state, even if it's just some electrons.
Vlatko Vedral: Yes. I think the main resistance actually in the early days, it's interesting that you mentioned this point, came from Niels Bohr himself. And Niels Bohr, according to many people almost one of the founding fathers of quantum mechanics actually. However, he thought that there is a certain kind of complementarity between being alive and even being quantum. So he thought about kind of quantum effects in chemistry being important, of course, for many natural processes. However, on the other hand, biology also relies on all sorts of organizational complexities as well, that are relevant for any living system. And he, in the 1930s actually, maybe around about the time of Schrödinger, was already speculating that you could actually experiment with the living system, but you may be able to only probe biological features. You will never be able to probe the kind of quantum chemical features that are relevant for, for living systems. So there were all sorts of ideas out there that living systems, if anything, may even collapse quantum effects, they may reduce quantum phenomena ultimately to some kind of classical physics.
and I think this, you started in the early days of quantum mechanics and, you know, they have persisted. In fact, even now some people believe that maybe living systems should ultimately somehow change the underlying physics.
So for us, this was really the key motivation to show that actually, probably this is not correct. And I think there is no contradiction between a full kind of quantum behavior and being alive.
Steve Hsu: Good. So you use the word collapse. And so let me just jump in there and say the question of putting some macroscopic system into a superposition state is sort of dual to the question of whether that macroscopic system can quote collapse wave functions. Because if the macroscopic system collapses the wave function, it will stop the process of itself being forced into a superposition state. So there they're sort of dual versions of the same question.
Vlatko Vedral: Yes.
Steve Hsu: Whether macroscopic systems can be placed into superposition states.
Vlatko Vedral: Yes.
Steve Hsu: And sorry, just to finish the thought. So already, just in our little story here already, by the late nineties, it was becoming early two thousands. It was becoming clear that having a large number of degrees of freedom would not automatically make you into a wave function collapsing system. At least some examples of macroscopic systems with millions of degrees of freedom could actually be placed in superposition states.
Vlatko Vedral: Yes. And I think that was really very important. Certainly one thing to immediately add to that is that it does become more and more complicated to maintain these superpositions and these large entangled states. And this is simply because you have to really keep track of all the relevant degrees of freedom.
You kind of, what you don't want to happen is that bits of the environment that you don't control properly, somehow start to interact with your superposition. In which case this would effectively, for all purposes, be like a noise classical noise that can disrupt, of course, quantum superposition and the more degrees of freedom you have, the more ways in which something can also be disrupted exists. And that's actually the key challenge for all of the experiments that we are discussing in this direction.
Steve Hsu: So let's focus in on the bug for a second. It's called a tardigrade and it's a fraction of a millimeter in size, and it can survive being cooled to super low temperatures so that you can isolate it from thermal effects in the environment.
Vlatko Vedral: Yes.
Steve Hsu: And so a good candidate system, perhaps for living beings, although in a sense that the state that it was in, when you did your experiment, it was kind of dormant, but, but then it was, it was brought back to life after the experiment.
Vlatko Vedral: Yes.
Steve Hsu: This thing then is a good candidate system because it's macroscopic, but it can be brought to a very low temperature and then isolated from the environment around it.
Vlatko Vedral: Yes, that's again the key feature and why we chose this particular system. As you said, it's incredibly robust. In fact I think biologists don't really understand yet properly how tardigrades managed to achieve this state of kind of suspended animation. You know, they almost switch off all of their relevant functions, biological functions, and they go into this dormant state. As you said, the state of TUN, I think. T-U-N, as it's known formally by biologists. They expel all the water of course. So they make sure to really withstand these extreme external conditions like low temperature and low pressure. What was interesting to us by the way, is that the temperatures that were relevant to us because the experiment was done with superconducting quantum bits. For these qubits to operate these qubits, they really have to be at very low temperatures. And when I say low, we are talking about 10 millikelvin between one and 10 millikelvin. So you know, this is something that's almost a million times lower than the room temperature. These are very fancy cryogenics. Very fancy fridges, if you like informally speaking, to actually get you this very low temperature.
So we were looking for a living system that could withstand that, but biologists didn't know whether they could withstand anything as low as that. Any experiments before that were things like blasting them off on a rocket into outer space. And I think uh, they, they in fact were completely okay with these conditions.
But now we are talking about, you know, roughly 3 Kelvin background temperature. So this is about a thousand times higher than anything that we did in our experiment. So we didn't know whether we could really uh, get them two temperatures as low as this. Of course, you know, evolutionary speaking, there is absolutely no reason why they should have this ability. You know, that it's still one big mystery. Very low temperatures and very low pressures as well. We are talking about pressures that are about comparably million times lower than the atmospheric pressure that, that, that we kind of normally live with.
So it's very interesting that the tardigrade went into this state of tun under these extreme conditions. And then as you said, when the experiment was completed, we could actually just put the tardigrade back into water and watch it revive and go about its own business.
You know, that was very interesting to see.
Steve Hsu: Yeah. So obviously we don't know of any reason why evolution would have over engineered this guy to be
Vlatko Vedral: Yes.
Steve Hsu: more robust than any environment that it could ever experience in the natural world.
Vlatko Vedral: Yes.
Steve Hsu: But let's, let's go back to the actual experiments. So for the audience to visualize it, you have an apparatus where you have two superconducting qubits, let's call them qubit A and qubit B. whole thing is at super low temperature.
Vlatko Vedral: Yes.
Steve Hsu: You sort of place the tardigrade in between the two, right?
Vlatko Vedral: Yeah. Yes. I think what you need to do um, this each qubit is a bit like a tiny electrical circuit. Maybe that's the best way to describe it at this level. And what you do have is you have a capacitor there. So you have two plates which are facing each other and which can periodically become charged and discharged due to the movement of the charges inside this circuit. And what we did is place the tardigrade exactly in the gap between these two plates. And the gap, actually, if you look at it, almost perfectly matches the size of the tardigrade. I think the size was about half a millimeter. And that's exactly it. It was lucky that this fits this gap between the capacitor plates.
Steve Hsu: Great. And I mean, so some of the reaction to the experiment when your paper came out, I remember there are some people who were skeptical saying, well, how do they know they really put the tardigrade itself into a superposition state? You know, actually all they did was modify the dielectric constant, between the capacitor plates. And we don't really know what happened inside the tardigrade. Maybe, maybe you could talk about that a little bit.
Vlatko Vedral: Yes, I'd love to talk about it because of course there are various degrees of confidence that you can have in these experiments. And certainly one has to bear in mind that these are the first steps in that direction. So I think the experiments could really be improved. Then I can actually comment on what we could do and what we are planning to do in, in, in that direction.
I think the idea is really to explain the mechanism by which the tardigrade gets entangled to the qubit. So what happens in a qubit, as we said, is that it has this charge that exists in the qubit, which actually exists in two different places at the same time.
It's, it's very similar to, in some sense, to the Schrödinger cat, where you have this atom, that's in two different states of having decayed that not decayed. Here, you have two different states where charge exists in one place and in another place simultaneously in this quantum superposition.
And now what you do is you actually introduce the tardigrade there, which then has to somehow respond to this ambiguous, if you like, quantumly ambiguous state of the charge. And what quantum mechanics says to us is that this generates two different states inside the tardigrade. Two different charges changed states inside of the molecules inside the tardigrade. Such that when the qubit is in one state of charge, the tardigrade is in, in one corresponding internal state. And if the qubit is in the other state, the tardigrade follows. And it's in another internal molecular state. So it's not quite dead and alive, if you like, it's not quite Schrödinger’s cat, it doesn't respond as extremely as that. But what it means really is that the microscopic state of charges inside the tardigrade, inside this living animal, have to respond in two different ways simultaneously to the superposed state of the qubit.
And that actually quantum mechanically means that they become entangled.
Steve Hsu: So this is good. This is getting into some details that I think only physicists would appreciate, but so it looked like in your paper, the model for the tardigrade for this specific calculation was some kind of big sum over dipoles. Is that right?
Vlatko Vedral: Yes. Yes.
Steve Hsu: And yeah, go ahead.
Vlatko Vedral: Yes, it's, it's really you can think of even this qubit and the superpose charge as some kind of induced dipole, which then further induces a response of probably many dipoles. In fact, this is one of the, one of the things that I mentioned we would like to do in the future is to understand better the exact extent of these dipoles inside the tardigrade. That's something we don't understand well, to the degree that we don't even understand how many dipoles exactly are involved. We don't understand the frequencies of these oscillations properly and so on. So I think that there are a lot of details that we would love to be able to understand, to improve our model. But yes, it was sufficient just to do an ordinary simple dipole dipole couples.
Steve Hsu: Yeah. So at an abstract level, I could fully accept your calculation because I would just, I was thinking well, okay. If the qubit is in a different charge state, surely there is some reaction from the, you know, degrees of freedom and the tardigrade. And so if, if the charges are in a superposition state than the actual guts of the tardigrade likely to end up in a corresponding superposition state, of which maybe we don't know the full details.
Then when I saw your sum over dipoles, I was thinking maybe these guys actually know something about the specific degrees of freedom. Like what, what the, are these specific molecules inside the tardigrade, or are they only effective degrees of freedom inside? So that part, I didn't really understand.
Vlatko Vedral: No. And I think it's a great question. And we don't understand that well in in a similar experiment that, that we did on a bacterium a while back, maybe some five years ago, we had a better understanding because a bacterium is a simpler organism and this particular one has been studied extensively because of its ability to do photosynthesis.
So I could tell you a little bit more about the energy level structures. I could even comment on exactly what kind of dipoles we're talking about, what kind of vibrations get involved in these dynamics. But I think with a tardigrade, we don't have that understanding. So you're right. This would be key, a key question because it would also tell us how microscopic this superposition ultimately is.
You know, what fraction of the tardigrade really does participate in this entangled state.
Steve Hsu: Yeah, I w again, like, I'm not an expert on any of this, but as I was thinking about it, I was reading your paper. I thought, well, surely whatever this dipole reaction in the tardigrade is, it's some collective mode and involves lots of, you know, individual electrons inside. So it is macroscopic in some sense, I can't avoid being macroscopic in some sense, I think.
Vlatko Vedral: I think so. And I think that's how we think about it. Again, this would be another suggestion for, for really further experiments, is that if you really want to demonstrate conclusively beyond reasonable doubt, the two systems are entangled, what we tend to do in, in the field is to separate these two system and to make measurements which are done independently on one of them on each of them without affecting the other system.
So they're kind of done locally close to the vicinity of where the system is and independently of the other system. And then somehow these measurements are jointly analyzed to conclude whether the state, the joint state, is entangled or not.
Here in this experiment. We were not able to do that because the tardigrade forms almost like a very tightly bound hybrid qubit with the qubit itself, with the superconducting qubits. And it's simply not feasible experimentally to make a separate measurement on the tardigrade to the measurements that you do on the superconducting qubit.
But I think if we were to improve our experiments and develop them more, this would be the direction I would like to go into to really probe more extensively the degrees of freedom of the tardigrade and to try to separate them somehow from the qubit and measure them independently.
Steve Hsu: Yes. So I think if people criticize your experiment and just say, well, we'd like it better. If they could take this last step, I think that's totally reasonable.
Vlatko Vedral: It's reasonable.
Steve Hsu: I thought I saw comments, maybe it's cause I was looking on that terrible place called Twitter, but, but I thought I saw comments where people just doubted that you had even made a macroscopic superposition state me reading your paper, it seems like that's probably the most likely thing that happened. Not not that even though you don't know all the details of it
Vlatko Vedral: Of course., I think of it like that too. And maybe another exciting thing to add. And you mentioned it briefly at the beginning, there was another qubit involved. And so if you want to demonstrate that something is really a bonafide, quantum system, then the ultimate test is whether you can really entangle that system to another genuine quantum system.
So on the one hand, we have the study grid inside, one of the qubits, forming this kind of hybrid macroscopic superposition, and then you have these extra superconducting qubits that now gets entangled to this first hybrid qubit. And in fact, the experiment verified with a very high fidelity that this state was entangled.
And in fact, I think they created all sorts of different entangled states. They demonstrated that you can do the full range of superpositions. So to me, I think this interpretation that it is a microscopic superposition also seems much more likely than anything else.
Steve Hsu: Right now, let's maybe fast forward and say 10 years go by and we're meeting again in some pub at Oxford or something. And, you know, I say, well, congratulations, Vlatko, you guys did this and this with a bacteria. And you did this with another tardigrade system. Surely now no one can doubt that biological systems, i.e. these examples we just quoted, can be placed into a superposition state. And furthermore, that means that from the Bohr-ian perspective, they're not things which automatically quote, caused the wave function to collapse. It seems like that's where, you know, assuming that these experiments go the way I think they would go, the field collectively physicists should come to that conclusion. Do you think that's what will happen?
Vlatko Vedral: I think so. I think maybe the key thing will be to really try to do this with two or more living systems. But it seems to me, if you really demonstrate the entanglement at that level, that it is clearly beyond, beyond reasonable doubt that there is no complementarity. There's certainly no conflict as Bohr would put it between being alive and being, being in a superposition, being fully quantum mechanical.
I think that must be the conclusion.
Steve Hsu: Right. And so I think for us, you know, lazy theorists who just lie on our backs and think about things, but we don't have to actually build the experiments. We, you know, for us, it just seems obvious. This is where it's heading. Right?
Vlatko Vedral: And I agree with you. I think quantum mechanics has been going more and more into the macro domain, really. And we're identifying more phenomena where we can actually, not just create entanglement, but then detect it, manipulate it reliably. And I agree with you that this probably characterizes the next of our experimental efforts.
Steve Hsu: You know, I often meet experimentalists who, you know, they're busy doing real stuff, so they don't have time to think about, you know, I don't, I hate to use the word quasi philosophical aspects of quantum mechanics. they, they might, I might say to them, well, do you think, will, do you think that we'll ever put a human in a superposition state and they might say yes, or they might say no.
And I say, is it possible you're in a superposition state right now? And you know, they might say, yes, they might say no. But then I ask them, well, will you be able to build a quantum computer with you know 10 million qubits and that thing will be in a very complex superposition state. They, of course, yes, eventually we're going to do that.
Vlatko Vedral: Yes.
Steve Hsu: But there doesn't seem to be you know, don't seem to, to me, there's a very strong dichotomy there because if I believe in AI, And I think, oh, you could implement some AI algorithm inside your quantum computer in the future. You've effectively made a brain which could be put in a superposition state, right?
Vlatko Vedral: Yes.
Steve Hsu: Yeah, go ahead.
Vlatko Vedral: Yes, it's very nice that you mentioned this. I love this topic, actually, that you just started because it seems to me, in coming back to Schrodinger, that the most exciting thing would really be to put not a cat into this experiment, but either a human or AI.
And the reason for that is that what you really want to probe is whether this entity feels any different when they become superposed, when they join this entangled state.
And I think, you know, we are not really able, sadly, to communicate very fatefully with animals. Otherwise of course a cat could do it, if you could ask the cat a few questions and get a response.
But I think before we do it with a human, because with a human again, the complexity is enormous. It seems to me that the best bet is to have maybe a quantum computer with, you know, maybe even a hundred thousand or a million qubits that actually simulates artificial intelligence. And really subject this entity to some kind of superposition and then try to interact with it to show that there is no ambiguity there whatsoever. And it's possible that this entity really exists in both branches of the superposition and in each of these branches, it feels just as we feel when we see things very clearly. So I think this to me would be a fantastic experiment to try to perform.
Steve Hsu: Yeah. So I think that's where things are going. And of course, if they make a general purpose quantum computer of extremely high complexity, then obviously it could also implement classical machine learning, AI type algorithms.
And so should be able to make a superposition state of a thing, which does machine learning or does some kind of AI like information processing.
So, yeah, I think, I think that's where things are going to go. And in the blog post I wrote, after I read your paper, I referenced a very old, funny story about a theorist named Sidney Coleman. I don't know if you're familiar with him, but he
Vlatko Vedral: Yes.
Steve Hsu: particle theorist at Harvard who mainly did work in things like quantum field theory. But he did think a little bit about quantum foundations. And toward the end of his career, he used to give this talk called Quantum Mechanics: In Your Face.
Vlatko Vedral: Yes.
Steve Hsu: Where he would really try to convey to the audience the visceral strangeness of quantum mechanics. And one of the little stories he tells is about Gork the robot.
So there's a robot, which is an AI that works for the experimentalists in the lab. And then Sidney makes this little joke. He says, well, of course Gork is just a robot. He doesn't collapse wave functions. So he can be put into a superposition state and, you know, so he can look at the output of the experiment without collapsing the wave function. And then it's only when he reports back to his master, the experimentalists, that the wave function collapses.
And this little story is sort of a retelling of, I think, what people call Wigner's Friend in quantum foundations. So Sydney was just kind of recasting it in this kind of science fiction setting. But it just shows you how absurd the whole thing is. Like, if I implement Gork inside a quantum computer, surely Gork has to split into different superposition states, right?
Vlatko Vedral: I would think so. I think it's an inevitable conclusion if you really take quantum mechanics seriously and if you really apply it to that level of complexity. You know, you're effectively treating everything in the universe as being quantum mechanical.
I think Bryce DeWitt had an interesting phrase, which he would call it quantum totalitarian property. David DeutschF told me that. And it's precisely totalitarian in this respect that anything that interacts with a superposition must become superposed exactly this way.
So there is no way out, I mean, to be, to be fair, of course, to people who believe in possibly different interpretations, it is possible that we will find something interesting when we do these experiments with more complex entities, maybe with, you know, with, with the human beings, with artificial intelligence, it is possible that we are missing some ingredients that are important to let's say thinking or perception or consciousness or... I think this is all wide open.
But at the moment, if you ask me to bet on it, I would fully agree with what you are saying. That somehow there is no reason why these things should not just behave exactly the quantum mechanical way as we are discussing them.
Steve Hsu: Yes. I think Hawking used to say, maybe he has the pithiest formulation, he says if quantum mechanics applies every degree of freedom in the universe, then many worlds follows trivially. I think that's a quote from Hawking.
But it's still highly controversial in our field. And, I would say, in the part of the physics community that I'm in, which is, you know, cosmologists, people do quantum gravity string theory, I would say the majority of people would place their bet. I'm a hundred percent or 99% confident, but they place their bet on this kind of quantum totalitarian or many worlds kind of view. But whenever I go to a conference in your part of the world, in your part of the field, people are much more guarded maybe because they're closer to the action people are much, at least in my experience, much less likely to say, okay, I either believe this, or I don't believe this.
So what, what do you think?
Vlatko Vedral: Yes, I think with cosmology it's, it's clear, as you said, because we think of the universe as a, as a closed system. And then the only way to consistently treat it as such is really to apply quantum mechanics to all the degrees of freedom in the universe. I think when you go to the other end, you know, into atomic physics the people who do experiments with these tiny objects, for them what is relevant is of course the rest of the laboratory, which frequently all practical purposes can be treated completely classically.
So I think, I guess they are not, they haven't made up their mind simply because you could in your model account for all of your results frequently, by thinking of the large part of your laboratory, all of the equipment really classically you don't really have to quantize you like.
However, that doesn't mean that these things are not quantum because after all classical physics is just a very special case of quantum mechanics, quantum mechanics, of course reproduces the full classical behavior in some very special limit. So, you know, these two views are not even contradicting one another. It's just, it seems to me people who do atomic physics, quantum optics, and some subatomic they're frequently more reserved simply because we haven't tested larger and larger systems, but I think we're going in that direction. And I think the opinions will change with these more complex experiments.
Steve Hsu: Yeah, I'm really looking forward to that. I think that it's really a matter of just our technological capabilities. And if you and your collaborators continue to succeed with constructing these more and more complex superposition states, then I think gradually, maybe it'll start with younger physicists, but, but at some point people will realize, hey, the implication is very strong here now.
So we, we probably, we probably do live in a quantum multiverse or something with lots of entanglement, even at a macroscopic level.
Vlatko Vedral: Yes, I fully agree. I think it's the, it's the most consistent picture. Certainly consistent with everything we've observed. And even theoretically, you know, frequently mixing classical and quantum physics actually fails to address all sorts of other issues. It's simply logically inconsistent actually.
So somehow I think ultimately we will acknowledge that everything has to be treated quantum mechanical.
Steve Hsu: Yeah, I would say that again, this is my own opinion and not everybody agrees with me, but in my own study of quantum foundations, the only version that I can really formulate in a logically consistent way, in a full way, is the many worlds interpretation. And the other ones all have problems,
Vlatko Vedral: Yes.
Steve Hsu: problems.
So it doesn't mean many worlds is correct, but I agree.
Vlatko Vedral: I agree. Yeah. I agree. I think my guess would be that if something really happens with quantum mechanics, to the extent that we have to modify it, then it's going to be similar to what happened with classical physics, namely, whatever supersedes quantum mechanics will be something even more complex.
I think then quantum mechanics will become like a special case of this more general theory. But I think it's exceedingly unlikely that we're going to go back to classical physics. I think whatever comes next will be somehow even more correlated and entangled in some sense, rather than a return to the classical world.
Steve Hsu: Yes, I think you're right. You know, there, there are ideas out there involving quantum gravity and things like this,
Vlatko Vedral: Yes.
Steve Hsu: maybe supersede quantum mechanics. But it's very hard again. Now we're getting into a slightly technical discussion that maybe only theorists would like to hear, but it's very tough to deviate from linearity a little bit.
So it turns out if you allow even a little bit of linear deviation from linearity, which you would need for a dynamical collapse model or something like this, you immediately get tremendous amounts of non-locality and all kinds of problems. So, it's really, we're really stuck. I mean, it's very tough to imagine a theory that swallows quantum mechanics as a special case.
Vlatko Vedral: Yes. I agree. I agree. yeah, The question is where, where would this extension go? That's exactly the question. In which direction would you extend that. I fully agree with you that changing linearity has a drastic consequences and in a way we can almost rule out quite a lot of these extensions in that direction, precisely because of these inconsistencies.
Steve Hsu: Yes. So I think we're kind of like-minded, maybe, about how quantum foundations are going to evolve over time, pushed by this, you know, advancing quantum technology. I didn't know, before we started this interview you had some intellectual connection to David Deutsch. So if I knew that if I that I would've guessed, maybe you and I are like-minded on this question, but, let me shift gears and talk about practical stuff like quantum computers.
So I think a lot of people listening to this podcast are not really physicists, but are technologists who are super interested in quantum computing. And so, you know, I would, I think because there's now a lot of money involved and there are companies trying to raise money and selling actual products, which are a kind of quantum computer.
I think there's a lot of hype out there in terms of what's really possible. And so, you know, some basic questions like, oh, when will someone actually implement something like Shor's algorithm at a scale that actually matters.
Vlatko Vedral: Yes.
Steve Hsu: You know, do you have any sense personally of will that be 10 years or 50 years from now? What, what's your feeling on this?
Vlatko Vedral: I tend to be, I tend to be more optimistic simply because like I said there was a point at which in the late nineties, the field really started to develop in an exponential manner. You were really at that time surprised by the ingenuity and by how many qubit platforms are out there.
And then, maybe five or six years ago, you mentioned that some heavyweights started to get engaged on, on the industrial side. And now we have almost every major company running some kind of experiment in operation on quantum computers. I think it's difficult to predict because the platform we talked about, the superconducting qubits, they are natural because they already belong to this solid state domain where the semiconductor, you the conventional computers, rely on. And somehow it seems to me that the choice of, of companies like Google and Microsoft and IBM, the choice of superconductors is simply out of convenience because these are maybe the easiest to integrate with, with the rest of the solid state that would support this kind of computation. Because remember, qubits are just one part of this, but the supporting elements that you need to make a quantum computer really, for all practical purposes are really classical in many ways.
And this support comes from the solid state. So you gain a lot by doing that. But I think the fidelity of your qubits is frequently compromised with respect to, for instance, what you could do with cold atoms. I think if you're a purist, if you're a physicist, you would probably say it, it might be better to try to build these things bottom up and to take systems one by one that, you know, you can already control with hugely, with, with very high precision, in fact. So these are things you know, where you can do a gate with the fidelity of six nines, you know, 99.9999% efficiency.
And this is really enormous. This is something that superconducting qubits cannot do. So with superconductors, you know, you read the news that people have achieved a hundred qubits or sometimes even thousand qubits. What you have to bear in mind is that they are not really universal. They don't have a high enough fidelity that if we were to scale them up, you could really do Shor's algorithm with the high enough fidelity to really outperform anything classical.
At the other end, atomic qubits, they can have individually high fidelity, but then as we put more and more of them together, and they do actually decrease infidelity. So there we are talking about that most, maybe 10 quantum gates that we can now manipulate. And that, that really limits the kind of computation that we can do. We cannot do too many gates with these qubits. Maybe a hundred gates with, with 10 qubits, which is not much, you know, you cannot really do a useful algorithm with that.
But I think if I extrapolate this, you know, if I think that some 20 years ago, we barely had a single qubit of these implementations, and now we are talking almost approaching 50, maybe sometimes even hundred qubits of lower fidelity. I'd be more optimistic. I would imagine that within the next 10 years, we would really have a non-trivial computation that would definitely outperform anything that we can do classically even, you know, a network of classical computers could not compete with that.
Steve Hsu: So just to be precise though. So this milestone where a computations performed that really just couldn't, you couldn't imagine doing it with classical
Vlatko Vedral: Yes,
Steve Hsu: computers of the current day or
Vlatko Vedral: yes,
Steve Hsu: 10 years. That could be though a kind of quantum simulation where you're deducing some property of some quantum
Vlatko Vedral: yes,
Steve Hsu: that, you know, the size of the Hilbert space is just so huge
Vlatko Vedral: yes.
Steve Hsu: do a classically, but we do get some results, even though we have a very kind of noisy, perhaps noisy, set of qubits that we're using to do the calculation.
Vlatko Vedral: Yes.
Steve Hsu: That I think that sounds plausible, but, like doing, doing a Shor factorization of a number that actually matters right. For say you know, a blockchain or crypto security, that seems quite hard, right? I mean, you, you, you, you need to, even if you have like six nines, you probably need some additional error correction, right? Which, which, which causes some really brutal overhead, on the number of qubits you need and things like this. Do you think something like that is possible in the next 10 or 20 years?
Vlatko Vedral: I think so again, I'm optimistic. You are right. That there are all sorts of simulations that people are doing, which already are, are very hard with, with conventional computers. But, probably they don't yet constitute anything exciting in terms of algorithms. And even, even on the physics side, we're still not at the level where we could say, oh, here is a phenomenon that we don't really understand, in complex systems now by simulating it on a quantum computer, we can actually understand it. You know, there are lots of hopes that we can understand even more complex chemical processes in, in, in this way. I think even there we are not at this level where we could achieve this. I think we need many more qubits.
So in a way, it is similar to what you said about Shor's algorithm. And I think at the moment, the only way we can see how to do that by engaging error correction. I think you're right. Having stable qubits on their own without investing extra redundancy to correct for errors, it seems unlikely at present.
It's not excluded. You could think of technologies that could actually be robust and quantum mechanical, possibly even at room temperature. You know, physics does not prohibit that in any way. But it seems unlikely from this perspective that this will happen anytime soon. So I think the road will probably involve quantum error correction.
But even then, you know, with this, with this huge involvement experimentally, I am still on the optimistic side. I think if you say, you know, 20 years, it is possible that we can, that we can even get to the Shor's algorithm in that time. It will be fantastic. I think as well.
Steve Hsu: It would, it would also cause a huge uproar in all kinds of security, cryptosecurity, things like this.
Vlatko Vedral: I think this would be a huge application, you're right. This really would matter a lot in the real world as well.
Steve Hsu: So on the, obviously there's been tremendous experimental progress. I have kind of lost track of the quantum algorithm side of things. Are there quantum algorithms that give speed ups on things that we really care about that are as dramatic as say Shor's algorithm? Or is that still the best example?
Vlatko Vedral: I think that the best example may even be it's, so there are all sorts of variants of, of Shor's algorithm where people have applied this to a, to a range of problems, but it seems to me, the thing that will really be useful is Grover's search algorithm. That's another algorithm. It doesn't it doesn't achieve, as far as we can tell, any exponential speed up. It's only a quadratic speedup. But it seems to me that in practice, for instance, if you're talking about an element over a database with, with a million elements, you know, if you, if you say classically, you need on the order of million steps to, to, to get to the right element by simply inspecting the one by one, if you like. Then of course, you know, square root speed up means w within a thousand steps you could actually get quantum mechanically with a quantum computer. And it seems to me that all of these subroutines in the ultimate quantum computer that we'll all be using, hopefully one day, will be of this kind of the Grover type.
Maybe ultimately that is the most kind of ubiquitous application, even though it's not exponential. The exponential ones are more specialized. And of course there is a class of these problems that are very important to us and that are related to Shor's factorization. But I think the search will actually be the main application, ultimately.
Steve Hsu: I think technically, I mean, even, I think factorization, it doesn't really take exponential time. It probably takes high polynomial time. So even Shor is kind of a high polynomial speed up,
Vlatko Vedral: Yeah.
Steve Hsu: But, but should I be worried that the two main algorithms that are gonna quote pay for all this quantum computing research they were already invented right away at the beginning?
Vlatko Vedral: No, I think, I think there are a lot of, you're right. These are the major developments really that we have. There are many applications now, which actually are applied to all sorts of kinds of quantum mining, you know, data mining problems. They are variants, of course. Ultimately what you do is, if you're obviously applying, this ginormous superposition of many qubits to try to simultaneously probe all of these aspects and then interfere them to come up with the kind of final solution. So obviously these algorithms are all naturally within quantum mechanics related to one another. So I don't think there's any worry there.
And it seems to me that we develop our quantum computers, it seems to me that many more algorithms simply happen and come up as we develop our quantum computers. At the moment, people are simply busy worrying what can I do with, you know, with a small number of qubits? What can I do with 10 qubits? What can I do with 50 qubits? What can I do with 100 qubits? But I think as we manage to achieve larger and larger superpositions, I think all sorts of new applications that possibly we can't even foresee now will, will in fact happen.
I'm also optimistic about these developments. All you need to remember is, simply the development of classical computers and how little people could, you know, in the early days in the forties and fifties, how little people could even anticipate what was going to come with the development of personal computers, for instance.
And I think this kind of revolution will also take place with quantum computers.
Steve Hsu: Yeah, I hope you're right. I think the question, I mean, there's a kind of pessimistic scenario, not technologically pessimistic, but kind of algorithmically pessimistic, which is that, you're going to go to a lot of trouble to build these things, but the ways in which they actually, other than quantum simulation and quantum simulation has its own huge set of applications for just as you say, understanding chemistry or material science or whatever. Quantum Chromodynamics or something. But in terms of the pessimistic scenario regarding algorithms, or I guess what you'd call quantum complexity theory or something is, oh yeah, you're going to get these very special class of speed ups from quantum computers, but we've kind of already figured out what they are and the, you don't get, you're not going to get anything beyond that.
That seems possible to me, although it's perhaps ...
Vlatko Vedral: You're absolutely right. And in fact, there are computer scientists who come to mind who say that as you know, with the improvement of classical algorithms as well, that ultimately they will catch up. And in fact, the quantum advantage will, will really become minimal. In this case. It is a possible view. It seems unlikely to me actually.
Steve Hsu: yeah,
Vlatko Vedral: It is a possible view to maintain.
Steve Hsu: I mean, I have a colleague who has developed certain methods for fermions – spin one half particles – but using classical computers. And he's always saying that actually, yeah, the gap is not really that big between what you guys think you're going to do with your quantum simulations and what we can almost do now with just smarter classical algorithms.
But I, on the simulation side, I'm more optimistic because just the huge size of the Hilbert space is just a problem for classical computers. It's a problem that's directly addressed by the quantum computers. So I'm more optimistic there than anywhere else.
Vlatko Vedral: Yes. Almost certainly. I think these first big applications are more likely to be on the simulation side. And I agree, I think that was five months original motivation as well to say, you know, using quantum systems to simulate themselves is clearly more efficient than using classical bits to do the same thing. And I think that is the main drive at present.
Steve Hsu: Yeah, no, he got it right away.
Vlatko Vedral: Yes, I think so. That's it. was it. That's exactly it.
Steve Hsu: Yeah. Well, I really enjoyed this conversation. We've gone well over an hour now, so I should probably let you go. Are there any final comments you want to make before we finish?
Vlatko Vedral: No, no, no, not at all. Thanks a lot for, for asking me all of these questions. I think it's very exciting. And both of us like this direction, I think, which is really fantastic.
Steve Hsu: Yeah, I think, in some qualitative sense, you know, there's still mysteries in quantum mechanics and the experiments you're involved with are really pushing us toward the resolution of those really biggest of all mysteries actually. So I'm really excited about it.
Vlatko Vedral: Yeah. Me too. Yeah. Thank you. Yep.