Related content

Patrick 00:00

Welcome back to the Barclays Brief. Now, quantum computing is a topic that keeps coming up and it feels like a very good time to try to at least understand it, because people are saying it could change everything from medicine to cybersecurity. So it's really exciting. It's pretty complicated, and it's a little bit sci fi. I'm joined today by Laia Marin from our Thematic Research team to help me make sense of it and hopefully bring the rest of the listeners along too.

Laia, you've been writing a lot about quantum computing over the last few years. Thanks for being here on the Barclays Brief.

Laia 00:33

Thank you.

Patrick 00:34

So let's get going. Let's keep it simple for me. Start at the beginning. What is quantum computing?

Laia 00:40

Okay. That's a very easy answer. And without getting in too much technical details when it comes to quantum physics and quantum mechanics, because we'll probably have to rename this podcast because that would definitely not be brief.

Patrick 00:53

Okay. Yeah, we would have to call it the Barclays not so brief.

Laia 00:56

Think of, think about three different parts. What is it? What can it do? And also the different ways to build a quantum computer. So what it is based on quantum mechanics essentially quantum computers, instead of using bits like classical systems that can take either 0 or 1. They use qubits which can take either 0 or 1 or both at the same time. And because of these properties, the computational power essentially grows exponentially rather than linearly.

Now that said, one thing to bear in mind is quantum computers are still too prone to error, the error management side of things is quite complicated, and from an infrastructure point of view, it's also very difficult.

Some quantum computers need to be cooled at around absolute zero temperatures. Now that said, the application side of things is very, very exciting. So think of quantum computers as being able to do things that classical systems can't. Thinking about factorisation of large numbers. Thinking about unstructured data problems. AI is very good at structured data. Well, quantum computers are really good at unstructured data.

Think about quantum systems modelling. Think about optimisation problems. Very complex portfolio trading optimisation, supply chain optimisation, energy optimisation. Now, the last thing to leave you with as well is that there's different ways to build quantum computers. There's different ways to build qubits, different modalities. We're talking about the industry mostly approaching five types: superconducting, trapped-ions, neutral atoms, silicon spin and photonics, each with their pros and cons and essentially we don't know a which modality will win and commercialise faster.

Patrick 02:42

So what's the most common misconception with quantum computing?

Laia 02:45

Oh good question. I would say it's the fact that quantum computers are in fact hybrid machines. And when I say hybrid machines its that you need to think that for a quantum chip, a quantum processor, a QPU to be useful, it needs actually a classical processor. Either a CPU or a GPU, for example.

And this is because if you think about what a quantum computer looks like, you have the hardware layer where the quantum chips sits, and you also have the software layer. Now the entire software layer runs on classical systems. So this is why one of the important technological sort of milestones that the industry still needs to achieve is that integration with classical systems and also quantum systems.

Patrick 03:33

Okay. So you're saying that the the classical computer, so the the laptop on my desk at home - that isn't dying, but in the future it could be kind of linked up to a quantum computer?

Laia  03:43

Absolutely. And I think this is a fantastic point that you make, because people should think of quantum as being additive to classical systems. We're not talking about cannibalising. From an applications perspective we're talking about quantum computers being able to solve problems that classical systems can’t. And on the other side. You have the fact that quantum computers are hybrid machines in nature. So absolutely you're right, in the future, data centers will have a quantum processor and a classical processor working together.

Patrick 04:15

Okay, I guess the key question though is why is this relevant now? Because throughout my career, quantum computing always feels like it's another five years away. Which I guess is why I've never really properly engaged with it. But something's changed at the moment, which means that investors are kind of increasingly looking at it. I know you're increasingly busy talking about it. What's changed and why do you need to care about this today?

Laia 04:40

Yeah, and I love this question because I think this is a very important point to make. In terms of the catalyst and what we’ve seen in the industry sort of progress at the technological level we've seen a very interesting progress. But not only that, we're already seeing the first signs of first mover advantage coming through. And on top of that, if you look at the investments from sort of a government point of view, it has really dramatically increased year on year. So quantum is actually becoming the next tech sovereign priority for governments.

Now, if I go back to your question on timings, I think it's very important for listeners to understand that in the next 12, 24 months, this is when we expect to see the first signs of what's called quantum advantage coming through. So essentially, the moment when quantum computers demonstrate superiority over classical systems in specific but useful tasks.

So think of this as sort of the proof of concept. And we finalise the technological roadmaps of companies. And within this timeline in the next 1 to 2 years, is actually credible with the view that the true inflection point will be by 2029, 2030, when systems are targeting to reach fault tolerant levels. Essentially, reliable, error free quantum computers are able to solve any quantum algorithm.

Patrick 06:00

Right, so this quantum advantage moment is in the next year or two.

Laia 06:04

Yeah, absolutely.

Patrick 06:05

Whereas the bigger impact is going to come at the end of the decade. So, I know you love your role as a thematic analyst, but I'm going to promote you. Let's say you're the CEO or the CTO, the Chief Technology Officer of a big company.

Laia 06:18

Amazing.

Patrick 06:18

What would you be doing differently today because of quantum? And how would you prepare for this revolution?

Laia 06:26

Oh good question. So if I would be the CEO, imagining myself through the board meeting, sort of a chairing the meeting, I would definitely be asking for monthly updates, at least monthly updates to my CTO. And this is because, I mean, it's one of the things that I mentioned earlier but from my point of view, first mover advantages will be critical, especially if I would be a CEO of a financial company or a pharma company.

We have seen in the last 12 months very interesting examples of sort of quantum advantages coming through whether its portfolio optimisation, molecular modelling.

If I would be the CTO, definitely the first thing would be think about talent, because there is a little talent war emerging, that people should be mindful of. And the second bit, probably where would be spending most of my time is playing with quantum algorithms and playing with the software. This is where we have seen the slower progress actually if you compared to the progress seen at the hardware level. And it's because it's a bit of a chicken and egg problem because on one side, you have the hardware that is not powerful enough yet to test the software and the algorithms. So it's a bit of a chicken and egg problem. But definitely, definitely experimenting with algorithms. And you actually don't need to own a quantum computer to do that, because you can already access quantum computers through the cloud.

And then the last thing will probably be thinking about comparing the modalities and specifically assessing scalability, because in my view, that's what will dictate, let’s  say, which modality wins versus the other.

Patrick 08:03

Right. So talent, experimentation, scalability are key. Okay so stay in the role of Chief Technology Officer for a moment.

Laia 08:11

Okay.

Patrick 08:11

I'm going to be the CEO and I'm going to test you.

Laia 08:13

Okay.

Patrick 08:13

At that board meeting I say which qubit modality is winning the race at the moment?

Laia 08:19

So look we have made our own comparisons and sort of benchmarking exercise. And I would say that the winning modality today is a modality called trapped-ions. And the reason why we believe it's ahead in the race is because on one side, they actually have proven to have a number of qubits, useful qubits. It's called something, it's called logical qubits. But on the other side we've also seen very high levels of accuracy.

And this is actually measured by a metric that's called two-qubit gate modality. But without getting into much of the technical details, just think about very high levels of accuracy. We're talking about reaching 99.9%, which is key because that's when you can start implementing error correction software. And that's a very positive feedback loop. Now that's the answer to who's winning probably today.

The only thing that I'll leave you with is that I also think there's a dark horse race we think here, which is a modality called silicon spin. Instead of being based, you know, within the atom side it's actually based on electrons. And it's a very recent modality. We haven't seen that many KPIs being published from, from this modality yet.

However, they're able to build and to feed, let's say, millions of qubits into a single chip. So from a scalability point of view, this qubit modality remains a very interesting technology to follow in my view.

Patrick 09:45

Okay, so we definitely need to keep an eye on that, the silicon spin modality, the dark horse. Okay. Back to being a research analyst.

If you're listening to this show as an investor today, where does the opportunity sit? How can investors gain exposure to quantum computing? And also, are there really enough companies to invest in today?

Laia 10:05

Yeah. No, absolutely. So we've done a very big exercise, mapping sort of the entire value chain of quantum computing as an industry. We actually have identified more than 40 listed players so, yes, there's a lot of companies in the space already listed. So we think about sort of the value chain into four different buckets.

The first one is thinking about the quantum processors, which are the companies that are developing the quantum hardware, essentially the quantum chips. Then you would have also the suppliers, you could also think about companies exposed to the quantum chip manufacturing design process.

And finally, also you can think about broader ecosystem enablers that are helping the overall industry sort of progress. Now, it depends on how you want to gain exposure to these as an investor. If you care about gaining exposure as a pure players, you could essentially think about, gaining exposure to the quantum processors. Now, the thing to bear in mind is that because we said we don't know which qubit modality will win today, in my view, if you choose one single company that is exposed to one single qubit modality, you have a very high technological risk that you need to bear in mind.

Now, if you want to hedge against that technological risk, you could then again probably go further down the value chain. Again, think of the suppliers. Think of the enablers, you know, bearing in mind that you lose that, exposure to purity from a revenue point of view, of course.

Patrick 11:37

Okay. So clearly there's loads of different ways to play the theme. It matters today. And the development in this technology is incredibly fast. So Laia thank you so much for being here and explaining it.

Laia 11:48

Yeah. Thank you for having me. Hopefully that I can make you all experts in quantum computing in just ten minutes. But, yeah, I think we got there. Thank you so much for having me.

Patrick 11:58

Okay, so there's a lot to digest from this conversation with Laia. But the key message is it matters now. Clearly the quantum computing systems are far from perfect and they're not scalable. But that quantum advantage moment that Laia spoke about could be just around the corner. So companies and investors need to start thinking now about how to be positioned to leverage the opportunity that quantum computing could bring.

Thanks a lot for listening to the Barclays Brief. Please hit subscribe and we'll be back here at the same time next week.