Nobel Prize Conversations

Experiments that demonstrated that the bizarre properties of the quantum world can be made concrete in a system big enough to be held in your hand.

Listen to the 2025 physics laureates talk about their Nobel Prize-awarded work and life learnings in this podcast conversation with Adam Smith.

Podcast transcript

John Martinis: John generally ran a great lab and guided us really well. But the combination of the three of us was very special. I’ve been trying to recreate that the rest of my career because it was such an interesting time.

John Clarke: The three of us, we realised that we had done a couple of things that had never been done before. The fact that a macroscopic object means size of my thumb could have made quantum mechanics.

Adam Smith: So there we heard John Martinis and after him, John Clarke, talking about the magical time when in 1985, they, together with their co-laureate, Michel Devoret, observed quantum mechanics operating in an electrical circuit for the very first time.

Karin Svensson: When quantum mechanics comes into these discussions, I always find it magical and wondrous and strange…

Smith: And you panic.

Svensson: And I do panic a little bit, but I think they do a good job of sort of inviting us into this very strange world.

Smith: And the Nobel Prize in Physics often takes us into the realm of quantum mechanics, don’t they? So here we are again.

Svensson: Well, you spoke to John Martinis and John Clarke, and we’ll also hear from the third laureate, Michel Devoret, from an interview during Nobel Week in Stockholm. I’m Karin Svensson.

Smith: And I’m Adam Smith.

Svensson: And you’re listening to Nobel Prize Conversations. This podcast was produced in cooperation with Fundacion Ramon Areces. We begin with John Martinis reflecting on its first year as a Nobel Prize laureate.

MUSIC

Martinis: It’s fine. I would say, for the first month or two, it’s pretty intense. Now it’s a few events per week and it’s quite nice. I was in University of Washington this week and had a good time. Really enjoyed talking to the students. The whole trip was great, but the students really have a lot of questions to ask.

Smith: And the students must all wonder the same thing, which is how on earth you have a project as a graduate student, which is so good that 40 years later you get the Nobel Prize. What magical touch to choose a project that is that good.

Martinis: What surprises me right now, in retrospect, is that 10 other groups were trying to do this at the time. And what has happened is at the time, people kind of just accepted quantum mechanics and doing fundamental tests of quantum mechanics like Bell’s inequality, okay? Another Nobel Prize. That was interesting and noteworthy, but not as fundamental. I think because of the development of quantum computing and sensing and the like, people realised how much more important it is to do that. And I would say at the time, it was a noteworthy experiment and people paid attention to it. But it was only because of development of the quantum computer. And now what thousands of people are working on this. That’s the reason why it’s important. It aged well, put it that way.

MUSIC

Svensson: John Martinis grew up in California and did his Nobel Prize awarded research at UC Berkeley. He worked at Google, where he and his team built a quantum computer and demonstrated that it could outperform even the best classical computer, achieving quantum supremacy. He continues his work on quantum computers in the startup company, CoLabs, and it all began with a love of tinkering with electronics.

Martinis: When I was growing up, my dad was always building projects in the garage and doing things. I would help him and go and do little things. Then I got these electronic kits from Radio Shack and started building things, borrowed a friend’s oscilloscope and built some digital circuits. I remember in high school checking out a book from the library that told you how to design transistor circuits. There was a methodology and you’d use algebra and math to do that. I though that was the most amazing thing because it was the first time I was really using algebra mathematics to figure out how to build something. I remember the first time I got a circuit to work from that, okay? It’s really exciting.

Smith: I can imagine that, yes, because it’s also abstract at school and people tend not to explain why you have to do this.

Martinis: Yeah. I think it’s really nice to connect that with building. So for example, in Santa Barbara, there’s an engineering academy that I served on the board and helped and my kids went to. I really like it because in high school, you’re building things. You’re using these tools, but you’re also using math and using algebra and you’re using the concepts that you learned in, say, your physics class to build robots and things like that, which are really interesting. I go around talking about how to educate students. I would say if students can get this kind of practical experience in high school, because they have the time to develop in hobbies like this. When you’re in college, that gets a little bit harder. But in high school, that’s a good time. Having a hobby, building things, learning how to troubleshoot, that’s a really useful skill. My dad was a fireman, but he loved his job. He used to say, you don’t have to go to work if you love what you do, right? And I took that to heart and I found physics. And fortunately, I was good enough at physics to do it and good enough to get the funding and do the things that I’ve done. I love what I do and it’s fun. I’m married. We had children. I’m a Christian. So, there’s other things that are more important in my life. But it’s something I do all the time I can really enjoy. I stayed up late last night, for example. I just bought a 3D printer. I got up, stayed late trying to figure out how it works. I’m trying to build mock-ups of what our integrated circuit wafer assembly is going to look like. So I was up late last night figuring that out. I love this kind of thing.

MUSIC

Svensson: We often talk about physics, especially quantum mechanics, as this very theoretical thing, but it’s very much about building things, isn’t it?

Smith: Well, I suppose science in general is about the interplay between experiment, building things and theory. All science proceeds in fits and starts between those two. I suppose perhaps quantum physics, physics in general is more obvious, because the theories tend to be quite, how would you say? Far-reaching.

Svensson: Theoretical.

Smith: Theoretical. Then they need experimentation to back them up. So sometimes the theorists have a sort of a heyday, and then they need the experimentalists to come up with their big successes in order to show that they’re on the right path. I mean, I recall that Richard Feynman said something along the lines of, “It doesn’t matter how smart you are. If your theory doesn’t agree with the experiments, then your theory is wrong.” So you can’t go too far out on a limb theoretically before, you need to be grounded in experiment.

Svensson: But are theorists and experimentalists completely different types of people?

Smith: That’s an interesting question. Experimentalists often tend to have played around a lot when they were young with things like electronics or radio sets or making explosives. Often the same is true with theoreticians, but experimentalists may be like tinkering. They find that that’s where they get their joy. It does take a lot of tinkering to be a good experimentalist. I mean, some of them carry on very visibly enjoying that process. Thinking of a very nice laureate called Martin Perl, who discovered the tau lepton. He loved playing with Macano, or as they call it in the US, erector sets, when he was young. But that carried on into later life. When he was quite old, he was still trying to develop a Macano set for adults, if you like, using big pieces where you could make real things like vehicles and go-karts. And he got quite a long way down the production line before the insurance people stopped him and said, “You can’t possibly sell this to people because they might create things with which they’ll injure themselves.” So I think the word is tinkering.

Svensson: That sounds like someone who’s not forgotten how to play.

Smith: That’s very true. And as we know very nicely, so many scientists say that they’re not really working, they’re just playing.

Svensson: Well, like his co-laureate John Clarke also started out in electronics, which we’ll hear more about in a minute. But let’s introduce him first.

MUSIC

Svensson: John Clarke was born in Cambridge and got his doctorate there, studying under condensed mataphysics pioneer, Brian Pippard. Clark continued his research career at UC Berkeley, where he bridged the gap between fundamental quantum mechanics and practical quantum technology. Like John Martinis, he also found his calling early in life and credits his time at the historic Purse School.

MUSIC

Clarke: I was 12 years at the Purse and probably more than anything else in my life that decided what I was gonna do for various reasons. I think there was one event in particular that had a huge impact on me. It wasn’t just about physics. It was more about electronics. At my father’s suggestion, I had several apprenticeships in the summer. And I had one in particular with a gentleman called John Barron. He had just started a company in electronics. And in those days, of course, electronics was only just beginning. But anyway, I was going to spend the summer working with him, and I wasn’t quite sure what was going to be involved. But I’ve always, as a younger man, I’d always been very interested in electronics and how things work. So he sat me down when I arrived, and he said, “Well, first of all, I need to tell you something about electronics.” He told me various things for an hour or two, and all of which I think I basically knew and understood. He said, “Well, to make any more progress in our discussion, you need to know about calculus. And I assume you don’t know calculus.” And I said, “Of course I know calculus.” And I’d been at the Purse School. And for the last three or four years, we’d studied calculus because that’s what you did. It was an incredibly good school. Founded, by the way, in 1615.

Smith: Predates the Royal Society.

Clarke: Oh, yes, it does. It predates practically every organisation except my college. So he explained all this material, and I got fascinated by this. And I spent the summer building various circuits that he suggested to me. It all worked out very well. But then when I got home, I had a workshop at home, and I’d always been very interested in electronics. And so to cut a long story short, I decided I would build an analog computer. It worked remarkably very well. It was all, of course, based on vacuum tubes. Transistors weren’t around in those days. And so I built it, and in order to see how it worked, I had to build an oscilloscope. So I built an oscilloscope, and that also worked very well. I remember sort of scrounging these various components I needed from various places. But anyway, in the end, I had this analog computer, and I could, for example, solve first and second order differential equations on it. In a sense, that’s what got me going.

Smith: I can quite see why. For a start, the confidence to be able to build things like that and to understand the equipment you’re using so deeply gives you confidence in the results you get out of it. And, yes, it’s a very firm foundation for experimental work.

Clarke: But I think that’s really what got me going. There’s one other little story I should mention, that the school, the Purse school, had just moved to a new building in Cambridge. If you are a prestigious school in those days, it should be opened by a member of the royal family. And so after some long discussion with the royal family, it was decided that it would be opened by Princess Alexandra. So she came to open the school, and one of her requests was that she wanted to meet with a boy from each of the disciplines that was being studied at the school. So I represented physics. So that’s another key to how I ended up doing physics. It was decided that I would in fact talk to Her Royal Highness about my analog computer. It was something like a 15-minute discussion. And I explained to her about what a computer is and how it can be used to solve, for example, differential equations. Then I had built this oscilloscope so she could see the results of the calculations. She listened very intensely. I mean, it was a wonderful photograph of the two of us sitting there together with my physics teacher, David Weber, who was sort of in charge of this particular discussion. At the end of it, the very large audience applauded. Then David Weber asked her, “Royal Highness, do you have any questions for John?” And she looked at me and she said, “You didn’t really expect me to understand any of this, did you?” And there was a sort of big gasp from the audience because that wasn’t what they expected. So I thought for a second and I replied, “No, no, of course not.” And so the whole audience collapsed in laughter. That was how I met this beautiful lady who was two years older than me. But it really sort of had a big impact on my career. I think this was one of the incidents that got me going about electronics.

MUSIC

Svensson: Growing up on the outskirts of Paris, young Michel Devoret used to build and launch homemade rockets as a hobby. Having spent most of his academic career in the United States, he has become a prolific inventor within the realm of quantum technology. He’s also taught a course on physics in cinema. As a child, he though physicists were cooler than other scientists.

Devoret: I got interested in science very early, thanks to books that were given to me from Christmas and birthday from, by family members. You have to think about that period. So this where we were in the midst of the Sputnik era. And the US then created a lot of books for children. Some of them were translated in French. There were excellent popularisation, books about science. I loved reading them. In those books, there were stories about physics and physicist. Something that struck me was that unlike the chemists, physicists were never in lab coats. They were dressed in everyday clothes that seemed to me a good omen. They seemed more free-spirited than the rest of scientists. And, on the other hand, I like to worked with my hands to tinker with various subjects.

MUSIC

Smith: A lot of your work involves measurement, very precise measurement. Many Nobel Prize laureates in physics are involved in precision measurement. It’s very fundamental. But I’m not sure that that’s generally appreciated, that it sounds in a way like something that, yes, it’s a necessary part of doing experimentation, but being better and better and better at measurement doesn’t strike people as particularly the cutting edge. But it absolutely is, isn’t it?

Clarke: It is, yes. I agree with what you just said, but I think that the scientists do this, appreciate this. I mean, if someone makes some device that makes a better measurement that had been made before, we all think, well, that was a really clever idea. But I agree with you. I think that there’s a general rule, people don’t necessarily appreciate the importance of getting the measurement right.

Martinis: I’m doing my own startup. We’re working very carefully about the culture of it. We have something called CoLab values. One of the statements is metrology, not mythology. Okay? Which kind of encapsulates a little bit where we have to be really careful about our measurements. And that there’s a lot of notions in quantum computing what’s good or people do things. But to really dive into, how does it work? Take the measurement. For example, I was at a conference yesterday and someone was talking about a new way to calibrate, and it was using Google DeepMind and AI. On the drive home, I figured out a really simple way to, at least to explain it. It may not be the way you want to totally implement it, but if you wanna understand how it works, I have a nice, simple way. And that’s kind of the metrology, simple measurements, understanding things very clearly.

MUSIC

Svensson: We’ll hear John Clarke talk about a device he made called the slug. What was it?

Smith: Well, it was a super sensitive vault meter that he made when he was a graduate student in Cambridge.

Svensson: And why was that special?

Smith: Well, it was based on the ideas of a previous PhD student in the same lab. So both John Clark and this other Nobel Prize laureate who preceded him as a PhD student, Brian Josephson, was supervised by a man called Brian Pippard in Cambridge. And Brian Josephson came up with a mathematical formulation to explain how current could tunnel through a barrier in an electric circuit. It was a thing that got named the Josephson effect after him. And basically, it was how an electron pair can display quantum mechanical behaviour and jump from one superconductor to another through a insulating barrier without having to climb the energy barrier of going through the resistant barrier. And that combination of two superconductors and a barrier became known as the Josephson Junction. It was a stunning piece of theoretical work for which Brian Josephson was then awarded the Nobel Prize. But what John Clark did was to take that Josephson junction and build one and use it to measure very, very small differences in voltage.

Svensson: So from theory to practice.

Smith: Absolutely. Yes, exactly. And quite fast.

Svensson: And why was it called the slug?

Smith: Well, I think John Clarke actually came up with a name afterwards which produced the acronym. He called it the Superconducting Low Inductance Undulatory Galvanometer. But I think that was created after the fact. The name actually arose. His supervisor though that’s what it looked like.

Svensson: Well, let’s get into it with John Clarke.

MUSIC

Clarke: I started to fiddle around with bits and pieces of niobium wire and blobs of niobium and various other things to try to make a Josephson Junction that didn’t depend on thin films. This eventually allowed me to make a device which involved a blob of solder, a big blob, several millimeters long, which was melted on a piece of niobium wire, which of course is also a super conductor. It turns out that the niobium wire has an oxide layer on it.

Smith: So that acted as the insulator.

Clarke: Precisely, yes. So anyway, I made this one evening and, I mean, liquid helium at the time was quite precious, and we didn’t necessarily have it every day. But anyway, I had some leftover from a previous experiment that day. And so I made this device, which was, by the way, at the suggestion of my fellow graduate student, research student Paul Raitt. And I made this device, which only took me a few minutes, and I cooled it down. And my goodness, there was my Josephson Tunnel Junction. So I took it out and I put it on my bench. Then Brian Pippard came around the next day. He was very conscientious. He would go to see each of his students every day, even if only for a minute or two, just to check in what they were doing. And he looked at this thing on my desk and he said, “Huh. It looks as though a slug crawled through the window and expired there last night.” And so the slug was born.

Smith: Your volt meter, the slug.

Clarke: The slug. Yes.

Smith: It must have been tremendously satisfying to have managed to do that.

Clarke: Oh, yes. The slug became known as the superconducting low inductance undulatory galvanometer. Because it wasn’t in itself a volt meter. You had to connect a resistance to it to make it into a vault meter. But in the end, it became a very sensitive volt meter. Brian Pippard’s project needed me to be able to measure 10 to minus 12 volts in one second. With this device, I could measure 10 to minus 13. So it was a five order of magnitude improvement on the state of the art at the time. This created a lot of interest.

MUSIC

Svensson: How did the three laureates meet?

Smith: Well, they came together in John Clarke’s lab in Berkeley in California. John Martinis was one of John Clarke’s PhD students. And then Michel Deverot arrived as a postdoctoral fellow. The three really hit it off experimentally. They had complimentary skills and just, it magically all came together.

Svensson: And there’s a fourth person who was instrumental to the discovery being made. Can you tell me about Anthony Leggett and his role in this?

Smith: Yes. Well, he was a physicist, particularly interested in the theory of superconductors and super fluids. He was awarded the Nobel Prize for explaining the very odd behaviour of a rare isotope of helium, helium three, at extremely low temperatures. He was among the first to explore the macroscopic realisation of quantum mechanics. So quantum mechanics happens at extraordinarily small scales. And the big question is, how do you get it to operate at larger scales? So human scales? And he was a pioneer in thinking about that. He’s a fascinating man because his first degree was actually in classics, in Latin and Greek. He had no high school experience in science, and, but he decided to do a second degree in physics. It was the Sputnik era. The world was receptive to people turning to science.

Svensson: It seems like a good way to approach the world to do classics and physics. You pretty much covered all bases, haven’t you, then?

Smith: Yeah. Well, certainly it seemed to stand him in good stead. He was extraordinarily influential.

Svensson: We’ll hear Michelle Devoret talk about Tony Leggett’s importance in his Nobel banquet speech. But first, let’s listen to all three laureates reminisce about the time in the lab at Berkeley.

MUSIC

Martinis: I took a class from John Clarke on condensed matter physics, which was great. I chatted to him and he was doing experiments on quantum noise in these electrical devices. My hobby is electronics. Quantum mechanics, of course, is very interesting. The combination of these two was really interesting and for other people too. But during my graduate student time, I went to a conference, I think it was down in LA, UCLA, and Leggett was talking about this macroscopic quantum time. I though it was the most interesting experiment around. Other people found it interesting too, but I guess I was a little bit more obsessed about it than other people.

Smith: And also, I suppose, you had the combination of technical skills and interests that were required. It was the combination of people. It was John Clarke, you and Michel Devoret, and somehow the three of you was a magic combination.

Martinis: Very good collaboration. I think we each brought our own skills. I mean, my skillset is electronics and understanding how to do. And again, it’s electronic device. Michel, of course, brought in the knowledge of dilution refrigerators and NMR and understanding the quantum system. And John, you know, generally ran a great lab and guided us really well. But the combination of the three of us was very special. I’ve been trying to recreate that the rest of my career because it was such an interesting time to work together and collaborate so well together. As a graduate student, you wanna have good mentors and you ought have a good experiment. Okay? And this really helped me start my professional development in a great way, just to really see how you should do an experiment.

Devoret: I went and worked with Professor John Clarke, and that was a very important example and mentor. He was wonderful. So he had a way of visiting us, John Martinis and I in the lab. He would come every day for half an hour, an hour and ask, “Well, what’s new today?” And he was really giving excellent advice because he was coming every day and asking if something was new. If we would find two new things one day, we would keep one and telling about only one thing and keep the other one for the next day, just in case we wouldn’t have anything great to say the next day.

Smith: Somehow the three of you worked together very well. It must be absolutely fabulous to see that happening in your lab and see people coming together and really grasping it.

Clarke: But it’s very important to understand how we came to work on this. And that was because of Tony Leggett who very, very sadly died two months ago, I think. He was also, of course, an Englishman. I didn’t know him in England, but I had met him at various meetings very well. In fact, I’d given a talk at University of Illinois Champaign-Urbana. And it was his idea. He asked this question, “Can macroscopic objects or macroscopic variables obey quantum mechanics?” And I think that the general response to that question was, “Well, no, of course not.” But it was very intriguing for Michel and John and myself. And I said, “Okay, guys, well, let’s set up and see if we can figure out how to do this.” It was a very challenging experiment. It needed incredible voltage sensitivity. And so I had a rather primitive dilution refrigerator at that time. So we built these sort of very early devices and cooled them down. I think we got down to temperatures as low as 18 millikile. And that’s really how we got started. After maybe three years of rather hard work, we published these two FSREV letters.

Smith: Yeah, those 1985 papers, yes.

Clarke: Yeah. One was the 7th of October and the other was the three weeks later, 28th of October.

Smith: May I ask you, because that question that Tony Leggett had posed, you were able to see that it was just at the limit of the achievable. You knew that you could perhaps get an answer to it, and it was about as difficult a thing as you could possibly tackle. Everybody in research is always trying to find a question like that, that is so bold that other people just think maybe it would be crazy to attempt. But you feel you have a way in, you feel you can get to it. What gave you the courage, in a way, to attempt it?

Clarke: Well, I think in a sense, I didn’t have that much to lose in the sense that I had a quite big research group and they were all doing various things that were going well. I always had the greatest respect for John Martinis and Michel Devoret, the fact that they were willing to go for this, because for them, it was more or less the beginning of their careers. And for example, John Martinis was in his first year as a research student. And Michel had just got his PhD from France. So for them, it was a big challenge. And I had always the greatest respect that they were willing to give this a go. And also, they had great skill that was never an issue.

Smith: I mean, obviously, you knew the importance of what you’d published in 1985. Did the world around you immediately catch on to its importance?

Clarke: No, I don’t think so. I think that it took a while. I think for the three of us, we realised that we had done a couple of things that had never been done before. The fact that a macroscopic object, I mean the size of my thumb, could have been quantum mechanics. And I think that It took people a long time to understand that and to understand the significance of it.

Martinis: Quantum mechanics is often described as the physics of a small, and that’s actually wrong. Quantum mechanics can do everything, even macroscopic things, you just have to get into the right parameters. So to show that quantum mechanics is more widely applicable turned out to be. Well, first of all, we could have shown that quantum mechanics wasn’t applicable to macroscopic, which would have been interesting. But then if you do show it, then you have these new devices, a new way to build quantum system. So to me, it was kind of a win-win situation. Either way, the answer was interesting. And it turns out the hold of quantum device and quantum computing thing was way more important that we realised at the time. But I would say physics has a very nice tradition of just doing fundamental tests. And this was talked about by Leggett, who came up with the idea. So we’re very much indebted to him and his thought here. And it turned out to be really interesting. But what we did in the experiment technically is we really had to figure out how to do the measurement and how to do the setup to show that it’s right. And by doing that in a very fundamental way, principled way, it then meant that subsequent experiments that were more and more complicated could happen well. And I think we kind of set the stage for people really understanding how this works. You know, one example, we merged microwave engineering the quantum mechanics so that you can use both languages to describe what you were doing, which is a very natural way to think about how to design it.

Devoret: Our experimental work stemmed from a question asked by Professor Antony Leggett, “Do microscopic variables, like for instance, the current in an electrical circuit, obey quantum mechanics? Not every physicist in the 1980s agreed this question deserved years of complex experiments. But as Tony said, receiving himself the Nobel Prize 22 years ago for his work on the super fluidity of helium three, “If there is something in the conventional wisdom that you don’t understand, worry away at it for as long as it takes. And don’t be deterred by the assurances of your fellow physicist that these questions are well understood.” In 2025, the 100th anniversary year of quantum mechanics, it is fair to say that some of the questions Tony was addressing to the foundations of quantum mechanics 40 years ago are still open. However, he clearly led the way to our experiment, which provided strong evidence that the macroscopic current in a tunnel Josephson Junction indeed obeys quantum mechanics.

MUSIC

Svensson: Adam, I’ve sort of been putting this off, but we need to talk about quantum mechanics. Can you explain the discovery that they were awarded for? The citation is “for the discovery of macroscopic quantum mechanical tunneling and energy quantisation in an electric circuit”.

Smith: Oh, well. Got it there. It is what it says on the tin.

Svensson: Well, hoping I need a little more help than that.

Smith: Quantum mechanical tunneling was something that, for instance, was proposed by Brian Josephson. It’s the way that a pair of electrons can borrow its way through an electrical barrier. So that’s a quantum mechanical effect, which is just beginning to transition away from things happening all within the atom to things happening at a slightly larger scale. A big question in quantum mechanics is why these things that happen that can be now observed to happen at the tiniest scale of individual particles don’t seem to manifest themselves in the real world. Quantum phenomenon like superposition where something can be in two states at once apparently clearly isn’t the case in the world around us. So this prize is for the demonstration of a quantum mechanical effect at the macroscopic scale, a scale that is a phrase I used earlier, more human. And in this case, it’s within an electrical circuit. So the three laureates demonstrated in an electrical circuit a quantum tunneling effect. So they demonstrated quantum mechanical phenomenon in something that is at a scale that you can work with. Electrical circuit. Then they also demonstrated energy quantisation, which is a essential quantum phenomenon. Quantum mechanics is named after these energy quanta, these discrete energy states, that things don’t exist in an energy continuum, but rather they exist in discrete energy levels. By sending microwave pulses into this electrical circuit, they were able to see quantized energy. This was another demonstration that they were observing quantum mechanical effects at this large scale, relatively large scale. And scale is what so much of quantum mechanics is about these days in the public discussion, because of course that’s the big question for quantum computing. Can you scale up from observing things at a tiny scale to a scale that is usable?

Svensson: But you need very specific circumstances to be able to do these experiments, right?

Smith: You certainly do. This doesn’t happen room temperature, you have to make things incredibly cold, very close to absolute zero. That temperature threshold for these effects is going up as the technology gets more advanced, but it’s still, everything has to be very cold and very isolated from interfering effects. So it’s all extremely delicate.

Svensson: And then this very sort of delicate experiments are supposed to be turned into everyday computers at some point. Can you explain that to me?

Smith: That is the big problem. You can use the power of quantum mechanics to store and retrieve information in a way that would be very much more powerful than the computers we have today. But in order to do that, you need to bring multiple huge numbers of quantum mechanical effects together in a single place. And that scale up is extremely tricky because of many things, but in part because of the conditions you have to impose on everything and the isolation of the system. And so people are working very hard to try and produce robust aggregations of these qubits, these quantum equivalents of the bits that we work with in computers all the time that will be the basis of the quantum computer. There are lots of people competing to try and find a way to build something that is robust enough to actually work at a scale where you can solve the sort of problems that it’s promised that quantum computers might be able to solve should they work.

Svensson: And what are those problems?

Smith: Well, one that gets talked about a lot and is worrying people is this RSA cryptography problem. RSA cryptography is the standard way of scrambling information so that other people can’t intercept it as you send it between two places. At the moment, that’s unbreakable. But it’s thought that a quantum computer that could be developed in some foreseeable time. Will be able to smash through that code. Then that’s a huge problem for everybody because it’s what everyone relies on to keep everything secret. So that’s the sort of near term problem that could be solved. Then the computing part in theory of quantum computing, we’re told is absolutely enormous.

Svensson: What are John Martinis and his company working on in terms of this? He talks about wafers.

Smith: Yeah. They’re trying to produce QPUs, basically. Quantum processing units. They’re the equivalent of CPUs, central processing units in our computers now, but for quantum computers. The wafers are the semiconductor slices that these are built on. So these things hold the qbits. They’re the equivalent of bits in our computers now. Qbits are quantum bits. Colab, their company, feel they have a way of formulating these that will be freer from errors and more reproducible than some of the competitors. But many are racing to try and produce the hardware for the quantum computer. And the big problem is scalability issue. How do you put it together and keep it functioning without interference or error?

Svensson: So you asked John Clarke and John Martinis about the timeline for quantum computers and got two very different answers.

MUSIC

Smith: One inevitably has to ask you since you’re so close to the forefront on this, how you feel the quantum computing story is going and how close we are.

Clarke: You know, I get asked that question probably three times a day. And I don’t know. I’m not quite sure where all that’s heading.

Smith: It would be nice to talk about the potential of quantum computing. It sits there as a potential as something which will transform things when it comes. Would you like to put a timeline on it coming and would you like to talk about what it will transform?

Martinis: First of all, I’d like to say that computing is such an important part of our lives now and technology. That if you can come up with a computer that has expanded capabilities, I like to say expanded instruction set over classical, then one should be working on this and figuring out and seeing if it’s good. And I think there’s practical applications. I think it’s kind of hard, but I think we can get there. But it’s a very much kind of science and technology worthwhile thing to do. In terms of timeline and the like, I’ve kind of shifted my point of view with a lot of other people because of the great results that’s been happening, new results, and understanding that with internet security and crypto security, one has to be careful about this. And right now, what I’m saying is quantum computer, general purpose, large scale, something like five to 10 years. People are getting serious about building the big machine now, which is good. I think a lot of that is just optimism building on each other and saying that we can do it too. I actually think it’s a lot harder than people think to do this. In our particular view, we embrace that hardness and saying, “Well, the way we’re gonna build it, even it’s so hard, is we’re gonna use the semiconductor ecosystem to manufacture the qubit wafers in order to do this and scale it up properly.” We’re trying to be a little bit more forward-thinking on how to do this.

Smith: So explain that to me a little. You’re going to use the semiconductor ecosystem.

Martinis: Right now, people are building their own clean rooms to make the chips for a superconducting quantum computer. But these processes are kind of based on 20 to 50-year-old technology. And people can do it, and people think they can scale it up. But when I look at the results, it just doesn’t look reliable enough. So what we’re doing is we’re working with tool manufacturers like Applied Materials and process knowledge companies like Western Digital to figure out how to build this using standard, scalable semiconductor tools. 300 millimeter tools where they’re building multi-billion transistor chips and to build it with a standard semiconductor and reliable process technology. Now, because you have to kind of change everything to do it this way, it’s taken us some time to figure that out. Things are looking good. We’re very optimistic. We can build it, but we have some ways to go, obviously. But the idea is if you can figure out how to build it this way, then you can use a standard semiconductor process line to just crank through wafers. That can be done in a very efficient, manufacturable way, and to get the reliability you want. Just to give you an example, right now, you make, let’s say, 100 qubits. Some of the qubits drop out, doesn’t work that right. But if you’re not getting 100% yield at 100 qubits or even a thousand qubits, how are you gonna do a million? Whereas semiconductor processing can print billions of transistors and have it work. And in particular, we’re making 20,000 qubit wafers. So we should be able to do 20,000 tunnel junctions, et cetera. But we have to be very serious in process engineering.

Smith: But if you get there, you’ve solved the scalability problem.

Martinis: That’s what we think. And it’s just not scalable. It’s also cost. If you do what you’re doing right now and scale it up, it’s in the tens of billions of dollars, maybe a little bit less, but it’s very expensive. Whereas if you manufacture with wafers, and we have to do wafer scale bump on it, there’s a lot of stuff we have to do. You can make it a lot less expensive and make it more reliable. And remember, with a regular wafer, you take a wafer and then you probe it at room temperature to see which ones work. If you have something that works at 20 millikelvin, you know, deep cryogenic, it’s very unlikely you’re gonna build a wafer prober to do that. So one has to think very carefully about the whole systems there. And we believe, in the end, it’s all about the reliability of doing this properly.

Smith: Listening to you, there’s a sort of lovely practicability about it all, it’s gotta work.

Martinis: It’s very unphysics-like. I say that in jest. I mean, the people who build, like, the big particle accelerators and other things, big science, I mean, they had to figure out how to do this. What we need to do is introduce that to the field where people are building their own clean rooms using old technology. We just really wanna think about that. But to be honest, frankly, what’s interesting about this is it’s super interesting to learn about how the semiconductor industry does things in this very sophisticated, high reliability way. For me, I always like to learn new technology and learn new things. This is just a fascinating thing to do.

Smith: When you talk about it, I can imagine you as a teenager talking about building transistors in the garage.

Martinis: Yes, that’s right. Or soldering things together. The way I like to look about it is, now I get to use multi-billion dollar tools to put this together.

MUSIC

Svensson: I want to ask you about mistakes in science. Are they to be avoided or embraced?

Smith: Well, embraced, definitely. No question.

Svensson: Why?

Smith: Well, Einstein said it. He said a person who never made a mistake, never tried anything new. And actually, now you ask that. Because we’re gonna be talking about John Clarke and Brian Josephson, I took down from the shelf Brian Pippard’s classical thermodynamics. He writes at the beginning of it, he writes this, he says, “Great ideas are more often arrived at by a combination of intuition and a judicious disregard of niceties than by a systematic and logical development of explicitly formulated premises.” And that, in a way, says it because you can’t kind of logically work your way towards new discoveries. Or you can, perhaps. But more often than not, that won’t get you there. What you have to do is make a leap. You have to use your intuition and ignore the niceties and take risks. That seems to be the way that things proceed. Those risks will often get you into hot water. I mean, there are different sorts of mistakes in science out there because there’s just the mistakes of using the wrong buffer or whatever. That’s also an important thing to do because it teaches you to use the right buffer. They’re the mistakes when you probe nature in the wrong way and it just won’t give you an answer. And you have to change your approach to get an answer. There are the mistakes when you think you know what’s going on and you ask nature a question and nature says, “Nope, it’s not that way.” Then you’ve asked it in the right way because it’s told you something that you didn’t know. I think mistakes generally are just to be celebrated. A lot of people talk about the value of publishing mistakes. That doesn’t happen very much. People tend to publish what worked. It’s hard to learn from other people’s mistakes. It’s hard to envisage how you would tell the world about your mistakes. So it’s not an easy thing to say, how do you publish failures? But there’s an interest in doing so. I mean, there are obvious examples like clinical trials where if you’ve done a clinical trial and something didn’t work, that’s quite useful to know. It tends not to get published. So it just tends to be buried. So somebody else might go off and do exactly the same thing, not knowing that you tried it and it failed. So it would’ve been useful to know.

Svensson: It’s sort of like always in life. It’s good to learn how to fess up.

Smith: Yeah, exactly. But in science, it’s quite difficult to know how that process would work. A journal of failures. It’s seriously talked about very much.

MUSIC

Svensson: Michel Deverot and John Martinis have some thoughts on this.

MUSIC

Devoret: There are two qualities which are essential. First of all, you have to be resistant to failures. You should not get discouraged when failure happen. You have to be able to to learn from your mistakes. Doing mistakes is important and you have to not to be discouraged by them. You have to learn as much as possible from them.

Martinis: This is how you learn. You build things, it doesn’t go right. And then you learn, and then you learn how to troubleshoot. The mistakes are actually the point at that point. Of course, when you get something that works, that’s great. I would say the other thing that I do is I call myself a realist. So I look at some technology and it’s kind of the metrology over mythology. I’m really critical. I try to be as self-critical as I can because you can easily delude yourself into thinking one way’s the right way to do it. I think sometimes people find it hard that I nitpick and talk about all the problems. We’re really balancing, we have optimistic plan, think it’s gonna work. On the other hand, we have to be very realistic about how things are working and how to do measurements and make sure we’re on track.

Smith: That’s a very important point and a very difficult thing to learn because necessarily you become very invested in your idea. You want your idea to succeed. To be self-critical is hard.

Martinis: Yeah. If you spend any time on social media, Twitter, whatever, it’s all about expressing your point. It’s not about pros and cons. So we actually have an active culture where you always have to talk about the pros and cons and balance it in some way. Yeah, we think that’s very important and just be very explicit about it. Let’s say someone in our project want to do something. I’m not gonna listen to it until it lists the pros and cons. Okay, that’s kind of step one. Then we can start talking about it. Because I really believe that’s important to think about it carefully that way.

Smith: You sound like you were kind of fired up from the beginning, that you were destined to become an experimentalist, an experimental scientist.

Martinis: Oh, yeah. That was absolutely the case. Now, you know, in my undergraduate, I did find in my classes, and I understood the concepts, but it’s really being an experimentalist, which is my natural skill. And again, it’s because I grew up building things and enjoy it. That just comes naturally to me. Like, managing a group, that’s harder. But I’m gonna say physicists are not maybe the best at communication. That’s not what we’re trained to do. That’s been hard for me.

MUSIC

Martinis: You really wanna try to pick things that would be very foundational and transformational. They don’t come often. A lot of the times you’re just doing good incremental work and developing the technology. And then every once in a while, you come across something that’s quite important.

Smith: And so with CoLab, your company now, you, you feel you have the possibility of creating qubits a little bit better than the competition.

Martinis: Yeah. I’m gonna say a lot better. But other people will probably say it’s incremental. What happens as we pitch this now, people are saying, “Nah, what we’re doing now is fine. Everything’s good.” And then I think once we show that it works and you don’t make good qubits really reliable, et cetera, then it’s gonna be, “Oh, yeah, we knew how to do this. Here’s this paper.” So that means you have a good idea. Okay, that’s my experience doing physics for many decades. But we have to show it. We have a good idea. I’m really excited about it. But unless you show that the technology is better, okay, it’s just an idea.

Smith: You look into very fundamental questions. For instance, you’re searching for the axion.

Clarke: Yes. I’m very active in that. In fact, I had a big meeting yesterday on campus about different approaches to doing that, which I’m gonna be very involved in, I hope. The axion is fascinating.

Smith: I suppose nowadays, you can go in so many directions with the technologies that you’ve developed. How do you decide what to work on? What sort of question appeals to you these days?

Clarke: I think part of the answer is that it is, you know, other people that I know and respect. And if they have a particular idea that I think I could contribute to, that’s always an incentive for me. But I think that working with other people has always been important for me. I like the personal interaction, but I also learn a lot from talking to them. I learn a lot from what they want to do with some particular device that I may have created. They say, “Well, if you could make it an order of magnitude more sensitive, then we could do this, which we can’t do now.” And I think that’s always a big challenge for me.

Smith: You’ve had successive achievements throughout your career. You’ve done things that kind of have given you a confidence to go up to the next level, whether it’s building a computer in your own little workshop at home or solving Brian Pippard’s challenge to make a more sensitive vault meter. It’s almost a sort of ladder progression of achievement giving you a confident basis on which to approach a bigger problem. I mean, I guess it just falls out that way, but it sounds quite structured.

Clarke: Yes, I was gonna say the same thing. Well, that’s what happened.

Smith: Yes. And maybe that’s the answer to all questions about looking back on a research program. Well, that’s what happened. I remember the chemist, or Bob Grubbs, who was a lovely, very humble fella at Caltech. I don’t know if you knew him. I remember he used to have a lovely phrase summing up that sort of thing. He said, “Well, we do our best, and we call it great.”

Clarke: I’ve never heard that before. I like that. Yes. Could I tell you one other little story?

Smith: Please. Yes.

Clarke: So we won the Nobel Prize with basically two short papers that we published in 1985. And the first one was on the 7th of October, 1985. And on the 7th of October, 2025, at two o’clock in the morning, I got this phone call which was from the Nobel Foundation. It was 40 years to the day since we wrote this first paper. I all thought that this was an incredible coincidence. It wasn’t something that they had figured out in any way. They make this particular call on Tuesdays. It turned out to be Tuesday, October the 7th. It was quite a remarkable coincidence.

Smith: There is a beautiful symmetry in that somehow one feels that something was at work in the background.

Clarke: Yes, that’s right.

Smith: And 40 years is quite a long time to wait, isn’t it?

Clarke: Oh, it’s a long time, yes. No, I had never imagined for a single moment that the three of us would get the Nobel Prize. When I got this call at 2:00 in the morning, my first reaction was it was just some kind of crank call. So I had to ask them a couple of questions to make sure that it was genuine.

Smith: That’s a beautiful sibetry. It’s a pleasure speaking to you. Thank you for making time for me.

Clarke: It’s a great pleasure to talk to, Adam. And I really enjoyed our conversation and tried to answer your questions. And hopefully something will come out of this.

Smith: It’s been a pleasure speaking to you. Thank you very much, indeed, John.

Martinis: Thank you. It’s been fun. I’ve really enjoyed it.

Svensson: So we made it to the quantum realm and back.

Smith: And we’re still in one piece. Excellent.

Svensson: I’ll probably meet you there soon again.

Smith: I think it’s unavoidable. It’s a story that just keeps on grabbing. Looking forward to it.