Transcript: A - Z of Tech Episode 19: Q is for Quantum
Louise Taggart:
Hello and welcome to the A-Z of tech podcast, and today’s episode, Q for quantum. As always, I am your co-host Louise alongside Shreya. To be honest, this is a topic that I do struggle to get my head around a little bit. I am hoping that today’s guest will be able to enlighten me somewhat.
I am delighted to say that we are joined on this episode by a trio of wonderful guests. We have Dr Alistair Brash, who is a postdoc researcher at the University of Sheffield, focussing on quantum optics in semiconductors. We have Sumair Zamir, who is a project manager in PwC’s emerging applications and technologies team. Last but not least, we have Rebecca Lee, who is a manager in PwC’s cybersecurity team, and who has just completed a master’s degree in information security with a focus on blind quantum computation.
Shreya Gopal:
Hi Louise, I am totally with you. Quantum is a term I've heard and used a lot, but never quite understood the context in which it's being used. I am really looking forward to hearing from our speakers this afternoon.
With that being said, Alistair, no pressure, but this feels like a great opportunity to ask you if you can explain to us what the term quantum actually means, for those of us who are not actually physicists?
Alistair Brash:
Thank you for the introduction. Quantum physics describes physics really at the smallest scales that we know of. Ultimately, everything in our universe is made of particles, and we call these fundamental particles, because we can’t break them down any further, they are the smallest unit. The most common examples would be the lightest, made up of particles called photons and electrical current is made up of particles called electrons. Quantum physics describes the behaviour of these individual particles. It’s worth saying that when we have lots of these particles, we don’t really see a quantum effect. For example, inside a computer every logic gate is actually operating on thousands of electrons, or in the fibreoptic networks that make up the internet, each pulse that represents a bit of data is actually thousands, if not millions of photons, but if we start going down to the single particle level, then we start seeing some very different physics emerge, and quantum technologies will be trying to exploit that to realise new technologies.
Shreya:
Wonderful, thanks for that overview, Alistair, it sounds like a lot of small particles make a very big topic called quantum. What does quantum physics have to do with other areas that we hear about, for example, quantum computing?
Alistair:
Quantum mechanics itself has been around for about 100 years now and it was initially very controversial. I would say, it took a long time before we had the experiments that could reach the single particle level, but steadily, throughout the previous century, we saw a lot of its predictions be verified by experiments. Then really more recently, as we confirmed the theory, we’ve turned our attention to what we could do to exploit this physics for technologies. I would say there are three main categories of quantum technologies. The first of these would be quantum computing, which is certainly in the news a lot at the moment. Quantum mechanics gives rise to a different logic, computational logic to classical computing. If a certain mathematical problem, fortunately some very useful ones are included in this category, they can give a very large improvement in speed.
Let’s take an extreme example, if you had a problem on the classical computer, that would take longer than the known lifetime of the universe, so 14 billion years. If we could make a perfect quantum computer, you could get this down to seconds. For certain problems, quantum computing could be much faster. The second of these will be quantum communication. This is normally focussed on security, because these are fundamental particles, we can’t split them any further, but there is also a fundamental law that says we can’t clone single particles either. What this means is that if you are communicating with single particles an eavesdropper has to actually take some of the particles in a way that disturbs the signal. Then the receiver is aware of this disturbance and they know that their security has been compromised, so you can try resending the information again with a different encryption.
The third category would be sensing. Certain quantum states can basically allow you to measure physical quantities with more precision than would be allowed in classical physics. A couple of possible examples will be ultraprecise clocks. Maybe, people have heard of atomic clocks, so that’s one thing; or an example I’ve seen in the news recently is making very precise measurements of gravity, so you could use these for underground surveys on building sites.
Louise:
Thank you, Alistair, I actually really feel like I am learning things here. This is really nice for me personally. Could you tell us then, with that context in mind, how does this translate into the type of research that you are looking at yourself?
Alistair:
In Sheffield, you mentioned in the introduction that the topic was quantum optics and semiconductors. That’s looking at the interaction between light and matter at the quantum level, single photon, single electron. \one example, one experiment that we do is that we put some light in from a laser and this excites an electron, and then from the way that we structure our semiconductor samples, the electron is confined. Then when that electron loses the energy that it has gained from being excited, it will emit a single photon. Maybe, the subtlety here is, with the laser we started with classical light, but through the semiconductor, we’ve converted it into quantum light, so single photons.
That’s the basis for the experiments that we do. One thing we could do with that is that we can make, what we call a single photon source, which you could use for secure communication, or you could also use it to build a light-based quantum computer. Then, maybe, one kind of development of that is, if you use photons, the difficulty is that photons don’t really interact with each other, which is a good thing, because it makes them less susceptible to noise, but you can do something, where say one photon changes the state of the electron, and then the state of that electron controls the state of another photon. You can create logic gates between photons using the semiconductor as the intermediary. Hopefully, maybe, that gives a bit of a flavour of what we are doing in Sheffield.
Louise:
Thank you, at this point, I will bring in Sumair, if I may, if we can turn to you on this topic, it will be interesting to hear where your own interest in quantum has come from, and how that translates into some of the work you are doing with PwC?
Sumair Zamir:
Thank you, Louise, and thank you Alistair, that was really insightful. I feel like I’ve heard that many times, but every time I hear it, it always gets me excited about some of the application and things around quantum. For me specifically, I would like to define myself as someone who is bit of a tech enthusiast. I love reading about these different types of new technologies that are coming around the corner. A couple of years back, quantum computing specifically caught my eye while I was working in the disruptive innovation team at PwC. One of the key elements of our role is basically trying to understand some of the technologies that may significantly impact our industries in the next 5, 10, 15 years, but one of the really interesting things around quantum, especially around the commercial element of it, is that everyone appreciates, at least at a high level, that it is something that can make a real revolutionary change in the way that we tackle specific problems.
Just going back to what Alistair was saying, there are very specific problems that quantum computing, you’ll notice I focus on the quantum computing side of this which is where my focus is, about what quantum computing can solve.
Now, in terms of which kind of companies are looking into this, and who is involved, it’s all over the place. Governments and all the large technology players are thinking about quantum and what it can do. At the same time you have start-ups as well, both on the hardware side and the software side. It’s a very exciting space. And, I am sure over the next 5, 10, 15, who knows, maybe even longer, there will continue to be loads more interest in what quantum computing can bring.
Shreya:
Wonderful, thanks for the introduction, Sumair, what are some of the growth areas or investment opportunities that we are seeing with regards to quantum?
Sumair:
It’s a good question. Whenever we do speak to clients, at least initially, for the last couple of years at least, it has been more around just understanding the applications of what quantum computing can bring to them. The interesting thing is, it’s across all industries, like, whether you go from aviation, to transportation, to cars, to finance, to material science, to industry, everything, everyone is involved. Now, as with all investment advice, on the decision making process, but the number I would actually like to share, is maybe, how much governments have started investing into quantum computing. Now, a couple of years back, the total global levels of investment were around like $2 billion dollars, which is still a significant amount if you think about it. Probably, still now we are not really anywhere near close to getting the quantum computer that can actually do very many things, but even over the last couple of years, we are now closer to a number, which is 22 and a half billion of investment from governments all around the world, which is incredibly exciting. You can just imagine if governments are doing it, there is a lot of investment going all over the place.
Even in the private sector, you won’t see as big numbers, but even last year we’re talking about investments in a couple of hundred million around a start-up, which is actually focussed on photonic quantum computing, which is another, obviously, Alistair, mentioned there just before, which is a type of quantum computing. My expectations, and I am sure the wider industry is that it will continue to pique people’s interest, and the investment, I imagine, will continue to interest.
Shreya:
Wow there’s definitely lots of opportunity there for the industry with quantum. Could you give us some examples that are already, maybe, in use that our listeners, maybe, familiar with?
Sumair:
Sure, thanks Shreya. It’s a hard one, because the thing with quantum computing is that to really get the real applications for use, as of today, we really need the hardware to catchup and that in itself could take quite a bit of time, but what a lot of clients and what a lot of company is doing, and they’ve most likely spent quite a bit of money doing it, is figuring out which one of these applications are more shorter term in the next five-years’ time, and which one is the one they should be investing into seeing whether they can actually get some noticeable improvement in results through the quantum computer.
A couple of areas that we could think about that, people may have heard of. If you start with simulation, for example, you are talking about like, say, drug design in the pharma space. We could even be talking about Monte Carlo, for example, in the finance industry. These are things quite short term. On the other side you’ve got optimisation problems as well. This is thinking more around, let’s say, logistics and supply chain management, quantum could help around that. Microchip layout, think about semiconductors, very interesting, because the next stage of, what’s the most efficient way to structure the circuit. The other side is machine learning and AI. Alistair, mentioned just around probability and things, but just around sampling, for example, there is a load of different variety of use cases, it all depends on which industry you are focussed on.
Louise:
Brilliant, thank you, Sumair, really nice to hear some of those used cases, and hopefully put it into a bit of context for our listeners. You did mention cryptography there, so that seems like an excellent segway into introducing our third guest, Rebecca. We have already briefly mentioned quantum computing, could you tell us a little bit about your background and your own interest in quantum.
Rebecca Lee:
Yeah, thanks Louse, definitely. I many years ago did my undergrad in Physics, and I did some quantum mechanics as part of that. More recently I’ve just finished a master’s in information security, and what I was really keen to do was to just understand a little bit around what quantum was. There are some really big terms like postquantum cryptography, or quantum cryptography, and I just didn’t understand what those were. That was the beginning of my journey, was to really try and unpick what the difference is. I focused on an area of quantum cryptography, which is essentially using a quantum computer, which as Alistair and Sumair were saying, is a machine that can exploit quantum mathematics and quantum logic. If you took that machine, could you create a new type of cryptography that you can’t do with the computers of today, and what would that look like. That’s what I focused on in my master’s, but there is this other also really cool area of quantum that’s often in the news, called postquantum cryptography. That’s where really the greatest minds, I would say, are looking for new mathematics, just what maybe you’d call it normal mathematics, not quantum mathematics to base a lot of our cryptoprotocols on. Take the ones that we have currently and make them resistant to a quantum computer if and when it's built. Those broadly, quantum crypto, which in summary is taking a quantum computer and doing something new, and postquantum crypto, where we are taking what we already have and trying to make it resistant to a quantum computer. Those were the two different areas, that’s where my passion was, trying to understand what that big problem was, and go into the weeds a little bit with some protocols that I looked at.
Shreya:
Very exciting, Rebecca. Given your experience and work with cybersecurity, how does quantum crypto and quantum computing play with your work on cybersecurity?
Rebecca:
It’s a really interesting one, and it’s something in the news a lot, and it can often seem quite scary for businesses, but the short answer is, that yes there is a horizon threat there around quantum computer being built, and it potentially being able, if its built correctly, let’s say, to be able to break a lot of the crypto, that businesses and the internet, and the worldwide web has in place today. But really in the work that we do, and what we are seeing, is whilst there is a threat, there are ways that you can prepare for that threat. What it really comes down to actually is doing it maybe a little bit of, what maybe you would call crypto housekeeping now.
In real simple terms, that means, finding out what you have and where you have it. Do you know all of the different crypto controls that you have in place, do you know actually who owns them, who looks after them, who manages them, and can you put a plan in place to change them and update them should a quantum computer be built that can actually break them.
What we are seeing is working with clients to look at that problem and start, maybe, preparing for the future.
Louise:
Thank you, Rebecca, it sounds as though, maybe, things aren’t quite as pessimistic as sometimes the media might playout, but it is sensible for businesses to start thinking about their reliance on crypto, and how they use it, and are there improvements that they can make to make sure that, if in a post quantum world, encryption is less secure than it is currently, we have mitigations in place.
Rebecca:
Yes definitely, and what you are saying, and what Sumair and what Alistair pointed out, is there are so many amazing things that we will be able to do with a machine like a quantum computer that can exploit quantum logic. There is so much to be really excited about, but on the flipside, if it was built and if it was built correctly to be able to crack some of our current crypto controls, obviously there would be a threat there, but the basics of knowing what you have and what you haven’t, and having a plan in place is a really good starter for 10.
Shreya:
Lovely, thank you for that discussion, Rebecca. If we can bring all of you in, maybe, for a bit of jargon busting, there is a term called quantum supremacy that’s been heard quite often in the media, what does that actually mean, any of you volunteer to start us off?
Alistair:
Well, I can have a go at that. Quantum supremacy is in the news a lot at the moment, because we have some experiments that have been published recently that have claimed it. It is seen as a landmark on the way to quantum computing. Quantum supremacy would be when a quantum machine is able to solve a problem using a quantum algorithm that couldn’t feasibly be solved in a reasonable timescale using the best classical algorithm.
Louise:
Then there was another term that you mentioned Alistair earlier in the discussion, about a qubit, could you explain what a qubit is?
Alistair:
Yeah sure, qubits are the quantum equivalent of a bit in classical computing. In classical computing bits are a one or a zero, that’s how all your information is ultimately encoded in ones and zeroes. In quantum, it gets a little bit more complicated and people often talk about the idea that it can be both one and zero simultaneously. The way to explain this, the simplest case would be, we could have what we call a superposition state, which is where it has got some combination of one and zero. The simpler superposition would be that you have 50% probability that it's 0 and 50% probability that it's 1. If you measure a qubit 100 times in that state, 50 times you get 0, 50 times you get 1, or well, if you measure it enough times, you get good statistics, anyway.
You could think of that state as being like tossing a coin, that’s the key bit. You can see from that, that basically qubits contain more information than a classical bit, and that’s the foundation of quantum computing that there is this extra information in there that you can exploit potentially.
Louise:
That was really insightful, thank you, sorry Rebecca, would you like to add?
Rebecca:
Yeah, I was just going to jump in to say, this is something that I get asked a lot and I definitely had when I was starting my journey in my head. The way that I’ve just managed to calm that itch in my brain is to just think of, like Alistair was saying, a bit is either a zero or one, and really that’s a mathematical framework if you think about it. It’s us saying, if the electron is there, we are going to label it one, and if it’s not there we are going to label it zero. Actually, a qubit is just another mathematical framework that describes what the particle is doing. Whether it’s spinning up, you would call that a one; and whether it’s spinning down, you might call that a zero. All of the in-betweens, because particles do actually spin on their access, you can label that in a mathematical framework, and we call that a qubit. Going back to what Alistair was saying right at the beginning is that quantum mechanics is maths and we’ve built computers that can exploit that maths. In order to run those computers and make a machine, you need a new mathematical framework to describe the qubit, which is the equivalent of the bits. In the same way that a classical computer exploits Boolean logic and you can create this massive machine that does really cool things, that’s really at its very core, Boolean logic. With the quantum computer at its core, its quantum logic and you need a qubit as your mathematical fundamental framework.
Louise:
Thank you both, if we take a step up above the particle, and the bits, and the qubits, and think about some of the timelines involved in this type of technology and computing becoming more mainstream, how might this compare to other types of innovation that we’ve seen in the past few decades? Maybe, this is one, Sumair, you could start with.
Sumair:
Sure, thanks Louise, it’s a hard one. I would say that, even when I looked at it a few years back, there was a lot of talk about, maybe we will get some sort of quantum computer, or even achieve quantum supremacy in the next, like, 3 to 5 years. But just based on the reading I’ve seen, it really depends who you are asking. Some people are saying we might have an error free quantum computer in the next 5 years, obviously they have a vested interest, because they are actually building the quantum computer. Then you will have other people, who say, actually you know what, 10 to 15, it's a very wide range. The way I actually like to think about it, is actually going back to when we actually came up with classical computers all that time ago. That’s how I think about it, it’s that scale of innovation. We used to have these classical computers where we used to take up rooms, there used to be huge rooms, and they would do the most simple things that you can think of. That’s how I think, we are pretty much at the same stage, or maybe we are even a little bit earlier in that timeline, but it’s really exciting how things can move around so quickly. So, short answer hard to say, but very exciting time.
Shreya:
Given that this might take a while to come through or maybe it has already come through, what are some of the use cases that you all expect we might start to see in the next five years or so?
Sumair:
I can try and answer that one. So, there’s loads and it really depends on which industry you are talking about, as I mentioned before. I can list out a couple, just to get everyone a bit excited. You can have bidding strategies in the finance sector. Again, you could even do online marketing, if you think about it, that’s a really big piece around, in the minds of loads of people, like, as you get forward, how much you actually pay for data when people are searching and using internet. Obviously, all of that could change really quickly as well. For detection, patient diagnostics, network distribution around the communication side is one too, video compression, and things you couldn’t even think about really. It’s like, it can be very technical. Even design optimisation around the industrial side. Just thinking about one of my favourite examples, wing design and airplanes, you can’t even think about it, but it's just simulating how does the wing actually manage air resistance. These little things, because you can get that potential accuracy through a quantum algorithm. Quantum, to a certain degree, mimics the natural world. If you can get an understanding of that and build the algorithm to support it, your variety of use cases across all the industries is very exciting.
The ones that I mentioned were specifically around the 5 to 10 years. Some of them are even more exciting, those are the 10 plus years, we can get into that some other time.
Shreya:
Alistair, would you like to add anything?
Alistair:
Maybe, if I could follow on from that a bit. Sumair, focussed mainly on computing, and I would say computing is probably, maybe, overall the hardest quantum technology to realise. In communication, I would say that probably we are already seeing applications. There are a couple of companies making commercial quantum cryptography systems. It’s not that clear how many they’ve sold or who is using them, but certainly companies are evaluating them as part of their security portfolio going forwards.
Actually, one really nice example that I remember reading about a few years ago, and possibly the first commercial quantum technology is actually a quantum random number generator. This idea of having a probability distribution with a qubit. You know I said if you measured it a hundred times, you get 50 ones or 50 zeroes, but each individual measurement is completely random. Basically, you can use that concept to create truly random numbers, which is something that computers can’t really do, they can approximate randomness, but they can’t create true randomness. Those have actually been commercially available for at least 15 years.
Shreya:
There is a lot of convergence between academic research as well as commercial application today, but there is a lot more to look forward to in the quantum space in the next 5 to 10 years.
Thank you very much for joining us, Alistair, Rebecca, and Sumair, on today’s episode.
Thank you to our listeners as always. Don’t forget to subscribe to our series so you don’t miss our next episode, R for robotics.
