The Science Behind Quantum Computers

In this episode, award-winning quantum physicist and CEO of Nu Quantum, Dr Carmen Palacios-Berraquero discusses the science of quantum computing, the main technologies currently at play in the industry, and the benefits that our growing understanding of quantum physics will enable.

Transcript

Samantha Mabey: Welcome to Entrust Engage, an open forum for the most innovative leaders in security technology. I'm Samantha Mabey and I'm your host. On today's episode, we're going to continue our conversations around quantum computing in a post-quantum world, but look at it through a bit more of a scientific lens. I'm joined by the perfect person to help do this, award-winning quantum physicist and CEO of Nu Quantum, Dr. Carmen Palacios-Berraquero, welcome and thank you so much for taking the time to join us today.

Dr. Carmen Palacios-Berraquero: Hi, Samantha. Thanks. It's a pleasure to be here.

Samantha Mabey: So obviously a lot about what we're talking about is quantum computing. So to get started, I'd love to get a high level feel for the road to quantum computers that has happened so far, believing that it starts with quantum theory and that being applied to quantum technologies, one of which of course is quantum computers. Is that the right linkage of things?

Dr. Carmen Palacios-Berraquero: Sure. Yeah. So I guess the history of quantum computing goes back to around the eighties when people started to look at how to apply quantum theory to computing. Quantum theory appeared at the beginning of the century and throughout the eighties there were some ideas thrown around. In the nineties, we had Feynman proposing that you could use a quantum system to simulate a quantum system. And then we had the first quantum algorithms were developed like Shor's algorithm, which could actually use... harness these quantum information principles to really provide an exponential speed up on a certain calculation. In this case, factorization problem, which can be used to break things like RSA encryption. And from then on, obviously theorists kept advancing massively, but also experimental physicists started to find physical systems, so new hardware in which you could encode this quantum information and use it for computation. And that is... so we're still finding the best ways to do this.

So from the nineties on there have been more than seven or so physical systems like ions or atoms or photons in which we can encode quantum information and carry out quantum computations. And this has resulted in what we have today, which is this industrial race, like a space race of different kinds of hardware approaches and it's all kind of exponentially accelerating. So throughout the kind of early two thousands and kind of 2010, so on, quantum computing was still largely an academic endeavor. And then throughout the past seven, eight years, the leaders, the pioneers, the professors that kind of pioneered these first experiments on quantum hardware basically started to move into industry and started to create these startups.

In 2015, we started to see these first startups and more investment coming in. And it's all been accelerating to the point that in 2021 there's been over three billion invested into quantum computing, which is I think double or triple what was invested the previous year. And we've seen the first kind of large exits of these companies and we've seen obviously some big results and advancements like 2019 Google's result where they claimed quantum supremacy or quantum advantage. So it's been a relatively long history with a lot of acceleration in the past few years.

Samantha Mabey: Yeah, no kidding. And with that investment, I love the term, it's like the new space race because it's exactly what it sounds like it is. So of course being a digital security company, when we talk about quantum and quantum computing, we talk about the threat timeline. So specifically the timing of the quantum threat to public key cybersecurity and general consensus is that that's approximately 10 years away, give or take. So that said, a lot of reports have the quote when there will be a quantum computer powerful enough to break through traditional algorithms. So one that's powerful enough, that sort of suggests that quantum computers already exist today. Can you explain the current state of quantum computers?

Dr. Carmen Palacios-Berraquero: Yeah, sure. So I'd say at the moment there's five leading approaches, two of them, ions and superconducting, the kind of oldest technologies, the most well understood, higher TRL technologies that have the backing of the largest companies like IBM, Google, and now the large newcomers like IonQ or Honeywell. And those technologies are the ones that we are starting to see on the cloud that we can access via the cloud. Today, the numbers of qubits are small, but nevertheless it's allowing people to use and learn and train themselves.

There's the newer approaches like spins and silicon and photonics, which are meant to be more scalable. So they are meant to more easily deliver bigger machines, but they're also 10 to 20 years behind in the kind of roadmap. And so it's kind of more uncertain when we'll see those technologies deliver. And then there's... we've got neutral-atoms, which sits kind of in the middle and has seen a lot of popularity this year with several startups raising tens or hundreds of millions. So at the moment, the state of play is companies that claim anywhere between 20 and a hundred or 120 qubits, some of them being able to put a few of them available on the cloud and roadmaps for the higher TRL technologies that take us to about a thousand qubits and less kind of well published publicized roadmaps of the newer technologies that are supposed to take us beyond a thousand qubits, but in a more uncertain timeline.

Samantha Mabey: Okay. So they definitely exist today in some sense. And of course, you know, mentioned Google, of course, claiming quantum supremacy in 2019. So I guess I want to ask, what is quantum supremacy and when they claimed it, did they achieve it?

Dr. Carmen Palacios-Berraquero: Yeah, quantum supremacy or quantum advantage, as I prefer to call it was a term coined a few years ago. That is meant to say it's a moment where a quantum computer can solve a problem that is infeasible for a classical computer. So when Google claimed quantum supremacy in 2019, they kind of did a calculation of how long the same problem would take for the largest supercomputer on earth. And it was very, very... many thousands of years.

After that, there was a few publications that said that actually if you look at a way that a classical computer could have solved that problem more efficiently, it wouldn't take so long. In any case, I think the thing to understand is that that problem that was being solved very fast by a quantum computer and very slowly by a classical computer was a problem that has absolutely no interest in the real world, it has no application, it's just a made up problem. It's a calculation that a quantum computer is able to do faster. So I think people are really moving towards, at least in the kind of industrial landscape, people are moving towards the aim being a commercially useful problem.

Samantha Mabey: Yeah. Applying it to something real.

Dr. Carmen Palacios-Berraquero: Exactly.

Samantha Mabey: Yeah. Okay.

Dr. Carmen Palacios-Berraquero: Well how far away from that are we? I'm afraid it's, again, another billion dollar question.

Samantha Mabey: Yeah, of course. Well, considering the amount of investment that is going into building a quantum computer right now, and even Google saying they've achieved quantum supremacy, that was 2019, so obviously a few years away, does that mean anything in terms of the timeline being realized when we hear the general consensus is 10 years away, but this quantum supremacy was achieved in 2019, does that mean the timeline has been accelerated or was this just one milestone along the way?

Dr. Carmen Palacios-Berraquero: It's a difficult one. So again, the timeline to do what? So we have some roadmaps, say take IBM's roadmap takes us to a thousand qubits by I think 2025 or something like that. It's difficult to say where a thousand qubits... how useful a thousand qubits are going to be. The problem that is kind of plaguing the industry, there's essentially two problems. It's hard to scale these machines, it's hard to make them bigger. It's a very, very hard problem to go from 10 qubits to 11 qubits to 12 qubits. And that's why we see such slow progress. And then the other problem is that there's a lot of errors in the processors. And so people are proposing error correction schemes that are much, much more complicated and complex than classical error correction schemes, which means that it takes up many physical qubits to kind of encode one logical qubit.

And so in some of these machines, even if you manage to create one that is a thousand qubit machine, physical qubit machine, you have to take into account how many of those qubits you're using for error correction. And the ratios are really large and so maybe you only get a few logical qubits to actually perform reliable computations with. And so there's this big kind of debate in the industry where people talk about largely two regimes. So the kind of near term... kind of five to 10 years away NISQ regime, which is a noisy intermediate-scale quantum computer, which is around the thousands. And then there's this other regime which is beyond one million qubits. And so some people really think that because of error correction needs, we'll really only see the use of computers beyond one million qubits. And then there's the other half of the community that really thinks that the NISQ era will really produce a lot of commercial value.

Samantha Mabey: So you sort of mentioned, again, you're talking about maybe the problem that the Google computer solved maybe not being so useful, and you've talked about some commercial applications. So we've talked about the threats associated with post-quantum, but I'm hoping you can kind of walk us through the opposite and maybe we can have a look at some of the opportunities.

Dr. Carmen Palacios-Berraquero: There's so many opportunities that eventually quantum computers will open up. So the real... why we are in this field is not really to accelerate current computing. It's really to be able to crack intractable problems that are just not solvable at the moment. So opening up completely new applications, completely new markets, completely new industry, that's like the real promise of quantum computing. And in that category is things like material and drug design. So being able to simulate, say drugs, so for example, to be able to design drugs with very, very high accuracy or materials. And that has obviously a lot of implications for healthcare, but also for things like climate change. And then there's also the more near term applications that people talk about and they're mostly to do with optimization, so finance portfolio optimization, some acceleration or improvement of some machine learning algorithms and also kind of scientific computations. So simulating quantum systems, simulating physical systems. So those are the ones that people are targeting as a more near term application of quantum computing.

Samantha Mabey: Yeah, I've heard healthcare mentioned a lot, which of course makes sense. There's a lot to uncover there. I like the climate change angle, that's interesting. And of course financial services being I guess a low hanging fruit. So another topic I'd like to touch on is just some material I came across on quantum random number generator, which as I understand is being developed to encrypt things online. It's also said that the quantum world is the only place you can genuinely create a random number. So I'm curious first about the quantum random number generator, but then how does it differ from entropy and cryptography as we know it today?

Dr. Carmen Palacios-Berraquero: Yeah, so good question. There's been a lot of talk and debate around quantum number generators. So a quantum number generator uses the measurement of a quantum system and the unpredictability of the outcome of that measurement to generate random numbers or to generate entropy. This comes from one of the main propositions of quantum theory, which is that the outcome of a measurement is completely random, completely unpredictable, even if you can perform a million measurements and you will get a distribution of outcomes, but each individual outcome is completely random.

So people over the past, I guess decade or so, companies and academic groups have proposed to use this as a source of random numbers for cryptography. The difference with physically generated entropy is, well, that in classical mechanics, everything is predictable. Strictly speaking you could, if you had a lot of computing power and you knew exactly how your system works, you could predict the outcome of a classical random number generator.

Now in the real world, that is verging on impossible because you would need a perfect model of your system and you'd need a lot of computational power. Plus obviously the world of cryptography has developed ways in which you can apply certain mathematical tools on top of the outcomes of the physical random number generators to make that even more impossible to crack essentially. So there's an ongoing, I guess, debate of what do quantum random number generators add to random number generators? And lately there have been a few companies that are managing to achieve quantum random number generation rates that are really, really high. So this kind of starts being useful for applications, say like IoT or 5G. And so this maybe starts to become something that is useful, not necessarily because of its quantum nature, but more because of the performance.

Samantha Mabey: So we're seeing a lot of activity obviously in IoT and 5G. So is this something that... is it being tested right now for those spaces or is it actually being used?

Dr. Carmen Palacios-Berraquero: It's not really been widely adopted yet. So I would say it's still at the beginning of kind of-

Samantha Mabey: In its infancy. Yeah.

Dr. Carmen Palacios-Berraquero: So there are robust products out there, but I think it's difficult to insert a new product to do with cryptography and cyber security. Right?

Samantha Mabey: Yeah.

Dr. Carmen Palacios-Berraquero: So I think it might take a little while to get through the regulations, but also to really find that perfect use case... product use case fit, I think.

Samantha Mabey: Well even IoT and 5G are emerging, right, so I can imagine coupling the two together. Yeah, that's just a lot of new in those spaces. Would you be able to provide a bit of an overview on quantum computing as a service and what quantum internet are and how they're linked to quantum computing? So I saw them referred to as sort of parallel networks. Will they follow quantum computing or are they something different that's sort of happening in parallel?

Dr. Carmen Palacios-Berraquero: So they are quite different things. So quantum computing as a service and the quantum internet are very, very different things. So start with quantum computing as a service. So that is simply being able to access quantum computers via the cloud. So AWS, Amazon Braket hosts, I think it's now maybe five quantum computers and one can buy time on them via the cloud and run algorithms. Apparently there's a big queue and there's three kinds of machines already online at the moment with maybe five to 11 qubits. So generally, so these machines are not in a data center, we're not there yet, by no means. These machines are in the labs of these companies with a lot of people around taking care of them. And it will take a while for them to be able to survive in a data center alone. But in general, there is, I guess a view that one of the main ways in which people will be able to access quantum computers will be via the cloud rather than on the premises of say, large companies. But that is I guess, yet to be seen.

Samantha Mabey: So before, if you don't mind, I'd just love to jump in. And you said that there are some people who are accessing it now and there's queue access it. What kind of people or organizations are accessing it now? Do you know? For what purpose?

Dr. Carmen Palacios-Berraquero: Yeah, no, great question. So it's a mix of academics that makes that research say quantum algorithms and research kind of different parts, quantum information theory or are trying to develop, say software for quantum computers and algorithms and use cases for quantum computers. And then there's the kind of R and D departments of large companies of industry that are getting themselves used to using quantum computers and training themselves and starting to form teams within their organizations that are able to understand and use quantum computers. And using a small quantum computer is just a way to get that going. And then quantum companies, so quantum software and quantum algorithm companies themselves would use them as well. So there's a bit of a mix.

Samantha Mabey: So yeah, I didn't want to interrupt you. I think you were going to go to quantum internet next.

Dr. Carmen Palacios-Berraquero: Yeah, so the quantum internet, so one of the main resources, computational resources of quantum computers is this thing called entanglement, which is this kind of inextricable link between different qubits. So in a completely different way as current classical bits, they're all independent from each other in quantum, they're kind of intimately connected. And so that is one of the things that makes... that allows quantum computing to be so powerful. And this usually happens inside our computer. The quantum internet is essentially a network that distributes this entanglement between distant places and can kind of carry quantum information between distant points using photons because light is the only thing that we know that can carry quantum information over large distances. So this is kind of still very much in the realm of academia and it's kind of been spearheaded by University [inaudible 00:21:41], and it's still kind of unknown what the commercial application of the quantum internet will be.

But if we kind of take a step back from talking about long distance networking and kind of talk about short distance networking, then that is where Nu Quantum, my company, sees a very, very big opportunity, which is using photons using light to share entanglement between distant but not too distant things. So for example, between quantum computing modules. So I spoke earlier about how one of the main limitations of quantum computers is that it's very difficult to make them bigger and bigger, right?

So another thing that you could do instead of making them bigger and bigger is that you could connect them together, connect smaller modules together. So the only way to do this is using light in a completely analogous way as classical computers have a networking backbone that allows large computers to exist, super computers to exist. This is what Nu Quantum is proposing for quantum. So what this means is kind of making these computers emit photons that carry some of that quantum information and via those photons being able to extend the entanglement between different computing modules so that one could carry out multi-core quantum computation and like that, help on the roadmap to scale.

Samantha Mabey: Very interesting. Yeah, so I guess you said you don't know the commercial application of it just yet, so it's not like we'll be on the quantum internet in 10 years, but-

Dr. Carmen Palacios-Berraquero: I guess it's kind of a different realm. So the quantum internet talks about networks over hundreds of kilometers, and this is what we don't really know what the application will be, and this will be possible only in about a decade, right?

Samantha Mabey: Yeah. Yeah.

Dr. Carmen Palacios-Berraquero: But we do see a really, really big commercial application and opportunity in short distance networking between quantum computers, which is not... it's kind of the same, it's based on the same science, but it's not usually categorized as the quantum internet because it's short distance.

Samantha Mabey: All right. Well I believe we have reached our time for today. So thank you so much. That was a fascinating conversation. I hope it was just as fascinating for [inaudible 00:24:08], and it's been an absolute pleasure speaking with you today. So I'd like to thank you so much again for taking the time and having this conversation.

Dr. Carmen Palacios-Berraquero: Thank you. Absolutely. No, it's been a pleasure. Thanks a lot.

Samantha Mabey: Yeah, no problem. So that's it for today's podcast. Keep up with new episodes by following us on LinkedIn and Twitter using the links in the episode description. Thanks for listening to Entrust Engage.