The quantum revolution: The race to build a quantum computer

This is an audio transcript of the Tech Tonic podcast episode: ‘The quantum revolution: The race to build a quantum computer’

Madhumita Murgia
I’m Madhumita Murgia. We’re keen to hear more from our listeners about the show. So we’re running a survey which you can find at FT.com/techtonicsurvey, and you will be in with a chance to win a pair of Bose QuietComfort Earbuds. FT.com/techtonicsurvey.

Erik Lucero
So we’re now entering into kind of the larger, I would say, kind of our lab floor, right.

John Thornhill
In a nondescript collection of buildings in the Californian city of Santa Barbara, engineers are building a revolutionary new type of computer.

Erik Lucero
So welcome to what we believe is the future of computing.

John Thornhill
This is Erik Lucero. He’s a quantum engineer at the tech giant Google. And these buildings are the centre of Google’s mission to build quantum computers. How many do you have here?

Erik Lucero
It’s, last time I count, I think we’ve got about 20 in the room.

John Thornhill
Each computer is about the size of a fridge freezer.

Erik Lucero
I mean, the size of these systems, it’s kind of like two humans could hug it.

John Thornhill
They’re housed in metal cylinders, and inside the cylinders is a spectacular array of gold plates, wires and pipes suspended like a chandelier.

Erik Lucero
And then in the middle, kind of in the core of this cylinder that we’re staring at, is a really kind of beautiful structure of metals that are like shiny gold, some copper braids. From there we have wires that emerge and come down and basically connect to that motherboard with the quantum processor inside.

John Thornhill
The thing Erik is pointing out there at the bottom of the chandelier is a single and very special computer chip. It’s the bit where the mysterious features of quantum physics are harnessed to deliver unprecedented processing power. To the people building them, these golden chandeliers represent a coming revolution in computing. They say future generations of quantum computers like these will be able to completely transform industries such as drug development and logistics. They’ll be able to solve mathematical problems previously thought impossible, drive scientific breakthroughs and model the universe.

Erik Lucero
And you’re looking at, you know, a prototype of that system and how you can imagine now to think about how is this gonna grow when we go from, say, 100 qubits that we have today to a million physical qubits in the future.

John Thornhill
Companies like Google are betting on this quantum revolution becoming a reality. They’re confident they can build the quantum computers to make it happen. Question is, can they succeed?

[MUSIC PLAYING]

John Thornhill
This is Tech Tonic from the Financial Times. I’m John Thornhill.

Madhumita Murgia
And I’m Madhumita Murgia. Around the world, tech companies are building quantum computers, and they say they’re going to change everything. In this season we want to find out if that’s really true. And in this episode, the race to build a quantum computer. Are quantum computers really coming? And if so, when?

So, John, in the previous episode, we looked at how quantum computers might one day be used to break encryption on the internet. That could be catastrophic for our ability to keep information safe from hackers or hostile governments. But for the most part, those fears were about what might happen if a quantum computer powerful enough was built in the future. Because right now, that computer just doesn’t exist.

John Thornhill
That’s right. I mean, today we have small prototype quantum computers, but none that we know of that are big enough to do something really impressive, like crack encryption. But the people building them say they’re well on the way to creating such a computer that could do something like that. It’s just a matter of time until we get that. So I wanted to meet some of the people building these computers. In the US, it’s big tech companies like Google and IBM, even Amazon, that are doing it. And I wanted to find out how close they are to building a proper, fully functioning quantum computer. And the interesting thing is they aren’t really thinking about building these computers to break encryption. That was the big fear we talked about in the last episode. They say these quantum computers could be used to do some really great things and tackle some of the world’s biggest problems.

Krysta Svore
When we think about the challenges facing humanity, climate change, food production, right, helping solve some of the world’s hunger problems, thinking about more sustainable sources of energy, better batteries, cleaner water. These types of problems require a level of accuracy, a different type of computing, an advanced technology that we don’t have today.

John Thornhill
This is Krysta Svore. She’s in charge of quantum software at Microsoft, another of the big tech companies involved in quantum computing.

Krysta Svore
Quantum computing will enable us to solve these types of problems that impact all of us. That’s our north star. We want to solve some of the world’s hardest challenges with the quantum machine.

John Thornhill
The tech companies say that where quantum computers could really make a difference is in chemistry, in finding new, useful molecules. This is because predicting how complex molecules behave is really difficult. And beyond the power of classical computers today, the theory is that using quantum computers, you could model how different molecules work, and this would allow you to discover new drugs or find new materials to build better batteries, or tackle climate change by finding a molecule that can capture carbon in the air, or when it comes to food production, help you make better fertiliser.

Krysta Svore
Because what’s behind food production? Fertiliser production. What’s behind fertiliser production? Nitrogen fixation. Nitrogen fixation is the process of producing ammonia. So what we’d like to do is find a catalyst and find a process that improves on how we produce ammonia. Quantum computing can help here because it enables us to understand these catalytic reactions that could take otherwise a century.

John Thornhill
These are the ambitions being set by many of the people building quantum computers today. But like encryption, this is still all theoretical. Companies like Google and Microsoft don’t have quantum computers that are big enough to tackle these kinds of problems.

Krysta Svore
In terms of our quantum computer, we’re really focused on scaling up. So we’ve designed our quantum machine at the core to scale to a million physical qubits and beyond. That’s necessary to solve these problems that impact all of humanity. So we have to get to a million physical qubits.

John Thornhill
Qubits are the essential building blocks of quantum computers. A classical computer encodes information in bits, strings of ones and zeros. A quantum computer has qubits, bits that are in a quantum state. In order to tackle the big computational problems they’re supposed to solve, researchers estimate that quantum computers will probably need at least a million qubits. The quantum computers we know of today are nowhere near close to that. For example, Google’s latest quantum processor has just 72 qubits. The biggest known quantum computer in the world, developed by IBM, has 433. And scaling up a quantum computer from just a few dozen or a few hundred qubits to a million qubits is a huge technological challenge. That’s because qubits are notoriously fragile. They’re made of single subatomic particles in delicate quantum states, and keeping them stable in those quantum states is really hard.

I heard a Chinese quantum computer expert likening the challenge to lining up a whole line of kittens and making them all behave at the same time. (Laughter) Is that right?

Julie Love
So I think, you know, it’s not a terrible analogy, and I do like the kitten analogy, but it is, you know, it is a complex engineering challenge as we scale up to large numbers of qubits.

John Thornhill
Julie Love works at Microsoft’s quantum computing lab in Seattle. The problem is that qubits are sensitive to all kinds of noise from the environment. Even a tiny amount of heat can disrupt them, causing errors in their calculations. So quantum engineers like Julie have to keep the qubits cold — and not normal cold, deep-space kind of cold.

Julie Love
This is the sound of quantum computing, like in the cryogenic lab. We do these measurements at super low temperatures, much, much colder than deep space. We use specialised refrigeration technology called the dilution refrigerators to get to these ultra-low temperatures. Inside them is kind of a nesting-doll structure of shielding and more cooling power. We want to shield that from the room temperature environment and get that decreasing temperature as we go into the heart of these systems.

John Thornhill
But heat isn’t the only problem. Electrical signals, magnetic fields, even cosmic rays. Anything can affect the stability of the qubits’ quantum state and stop the computer working properly. Microsoft is developing a special type of qubit that is supposed to be more resistant to this environmental noise.

Krysta Svore
We’re taking a very different approach to our qubit, which is based on this topological qubit. When that’s inherently protected from noise in the environment, in a way that makes it really robust and resilient.

John Thornhill
Now you don’t have to worry about what a topological qubit is. The point here is that keeping qubits stable and working properly is the major challenge of building a quantum computer. And the more qubits you try to add to your computer, the bigger the problem becomes.

And you’re saying that you would need about a million qubits to do the really cool stuff. What’s your best guess about when we’re gonna get there?

Krysta Svore
You know, scientific breakthroughs don’t always come on a predictable timeline, but we’re looking at years and not decades for this level of innovation.

John Thornhill
Different tech companies and labs are trying out different approaches to building quantum computers. They use different materials to build their qubits, and they try different ways of cutting out the noise and reducing the computing errors that noise creates. But it’s not yet clear which approach, if any, will work. Microsoft says it’s years, not decades, away from a million qubits. As of now, it hasn’t built a computer with any qubits yet. Google, for its part, says it hopes to have a fully functioning quantum computer by 2030. The benefits of these computers, they say, are potentially massive. Solutions to climate change, new drugs, new materials — problems that used to take centuries, solved in minutes. But there’s another reason these companies are so keen to build a quantum computer. It’s an obvious one. There might also be a lot of money in it.

Tech companies might want to change the world, but if they succeed in conquering the fiendishly difficult engineering challenges involved in building quantum computers, those computers could have massive commercial value. If quantum computers can do what people say they can do, it’s hard to think of any industry that wouldn’t be interested in having one around. Investors are taking note.

David Cowan
(Notification sound) Yes. Hello.

John Thornhill
Hi, David. It’s John Thornhill here.

David Cowan
Hi John.

John Thornhill
Thank you so much for doing this interview.

David Cowan
Let me, I need a . . . I need about 30 seconds. I have to remove the cat from my room and retrieve my coffee.

John Thornhill
Very important.

David Cowan is a renowned venture capitalist. He works at a firm called Bessemer Venture Partners in Silicon Valley. He’s invested in a number of quantum start-ups and thinks that the potential commercial applications for quantum computers are huge.

David Cowan
I think that the first industries that will be most impacted are the ones involving logistics and the orchestration of logistics. And when we think about optimisation, that also includes financial trading applications. When we think about machine learning and the advances that we’re seeing around us in artificial intelligence, it’s hard to even fathom the leap in intelligence that we will see when quantum computers enter the fray. But when I think about how quantum computing will really impact humanity, by far the most profound impact will be on pharmaceuticals, on drug discovery. That will take maybe a little longer. But all of this kind of modelling can be done with quantum computers. So there are just so many different realms of computation that will be disrupted when quantum computers are commercially available.

John Thornhill
But in the future, the idea isn’t that every company that wants a quantum computer will have one installed in their office. Instead, the tech companies with the quantum computers will be able to charge other people for access to their quantum computing power. So getting to a million qubits might not just change the world, but make the tech company that builds them a lot of money.

David Cowan
It’s hard to even quantify the available market size for quantum computing. It would be like trying to quantify the market for computer transistors back in the early 20th century. It’s not just replacing supercomputers that we have today. It’s the idea that we can unlock solutions that we just couldn’t even dream of achieving in the past.

John Thornhill
Hi, Madhu.

Madhumita Murgia
Hi, John. So it sounds, from your reporting out in California for this episode, like there’s a lot of optimism amongst these tech companies about quantum computing.

John Thornhill
Yes, it sure is. I mean, I think it’s hard, when you meet all these quantum researchers, not to be infected by their optimism. They think they’re working on one of the world’s biggest challenges, and they’re really excited about that. And they’re also really excited about the possible uses of these quantum computers. They really do think that they’re going to be game-changing, but they clearly have to cross this valley of death before they get from where we are now to where they want to be in the future. And I think that is a genuinely exciting challenge for them as well. It would certainly get them out of bed in the morning.

Madhumita Murgia
Were you convinced, you know, did you, when you went around and reported and kind of saw some of this stuff in action, did you feel as confident as them?

John Thornhill
Huh. Well, I don’t think I’m as confident as they are. But I’m fascinated by this possibility that you could use classical computing in combination with nascent quantum computing, so-called kind of hybrid computing to get to where you want to get earlier than you would otherwise get. So I think that’s a really interesting field as well. That ah, the possibility of just mixing the strengths of both classical and quantum computing.

Madhumita Murgia
Yeah, the thing that really strikes me is how hard it is from the outside to judge progress in this field because that’s kind of defined differently by different types of people, isn’t it? And really, to work out how close we are to having that holy grail of a quantum computer that is scaled up enough to do anything useful because it sounds like they’re still miles away from that number of qubits, whatever that number may be that everyone says we’ll need. And so there seems to be this stark contrast between the hopes and optimism that people have for what it could do and really what it’s doing today.

John Thornhill
And in the meantime, even some of the recent apparent breakthroughs in quantum computing seem to be a bit unclear.

Madhumita Murgia
Yes, that’s right. Back in 2019, I broke the story about how Google reported achieving something called quantum supremacy. And that’s the idea that you have a quantum computer that can do something that a classical computer just cannot do.

News clips
Google announced today they’ve achieved quantum supremacy . . . The breakthrough here, the fundamental quantum computing potential and the way it could transform our lives, it makes this a very exciting day . . . 

Madhumita Murgia
And that was a big deal for the field because a quantum computer beat a classical computer for the first time. And with just a few, you know, 50-odd qubits.

News clip
It can perform a computation in seconds what would take the world’s fastest supercomputer thousands of years to do that same calculation.

Madhumita Murgia
But then it also turned out that the problem that it had cracked wasn’t a very useful one. It couldn’t really do anything. It’s an academic mathematical problem.

News clip
Some sceptics say Google’s overselling its achievement.

Madhumita Murgia
And then IBM jumped in and said that it wasn’t really true that it would take a classical computer a thousand years to do. So maybe it sounded like Google was exaggerating.

John Thornhill
But I guess the tech companies would say that even if we’re not there yet, the science is proven. So it’s not a question of if, but when. And you know, when it finally does arrive, those applications will be enormous.

Madhumita Murgia
But that’s the problem there, isn’t it? Because there’s not even really a consensus on how useful the quantum computer could be, what those applications will actually look like. For example, I spoke recently to a physics professor, ‪Sankar Das Sarma. He’s at the University of Maryland in the US and a veteran of quantum computing. He’s been at it for 25 years, and he was telling me that he’s really worried about the amount of hype that’s kind of built up around quantum computing in this moment.

‪Sankar Das Sarma
I had been involved in quantum computing research up, almost from the beginning. I am as pro-quantum computing as anybody can. I obviously want to see a product that’s called “quantum computer”. But on the other hand, wishing something is not gonna make it happen.

Madhumita Murgia
So he said that the main problem with the way that we’re talking about this all today is that we’re talking about these theoretical applications for what’s also a theoretical machine.

‪Sankar Das Sarma
This has been the great shortcoming of the field. We all thought in the beginning that there’ll be many, many uses for quantum computer. The computer scientists would figure it out. But it really hasn’t happened. You know, people talk about, oh, it can do chemical reactions faster, so maybe drug design will become easier. Maybe it can solve electronic structure, so fertilisers will become easier. But if you look at those usage and you know, the advantage is very small and it’s not even clear there is an advantage.

Madhumita Murgia
He made an interesting point when we spoke, which was that, you know, there’s just one real proven application for quantum computers if they’re built, that we actually know about, and that’s breaking encryption, which we’ve spoken about before. We covered that in our last episode. And if you built a good quantum computer, a working scaled-up one, then you can break encryption. And that’s something that’s actually been proven by mathematicians. But everything else that’s being talked about, you know, whether that’s drug development or finding molecules to fight climate change or new types of polymers, those just aren’t proven. We don’t know for sure that quantum computers will be able to do those things. We’re just hypothesising that they might, based on a pretty good hunch.

John Thornhill
Well, there’s clearly some debate about what quantum computers could do. But does he think that quantum computers are really coming?

Madhumita Murgia
So I asked him this, and his answer broadly is, yes, he believes in them, but just maybe not as soon as the tech companies say they’re going to.

‪Sankar Das Sarma
Is a quantum computer inevitable? My answer is yes, there will be a quantum computer. I can guarantee. I may not be around to see it, but there’ll be a quantum computer. But, you know, I mean, I want to put a slight dose of reality in it. It could be 100 years.

[MUSIC PLAYING]

John Thornhill
On the next episode in this season of Tech Tonic: how does a quantum computer really work?

Unnamed speaker
In the quantum world, what you’re trying to do is get that ability to control nature at that tiny, minuscule, microscopic scale that allows you to access those quantum states.

John Thornhill
We dive into the weirdness of the quantum world to find out how quantum mechanics is changing computing, even if we don’t really understand it.

Unnamed speaker
Quantum mechanics says that’s what you see when you look at the world. But that’s not what really exists when you’re not looking at the world. Surely you don’t really believe this. That’s just crazy.

John Thornhill
This has been Tech Tonic from the Financial Times. I’m John Thornhill.

Madhumita Murgia
I’m Madhumita Murgia. Our producer is Josh Gabert-Doyon, senior producer, Edwin Lane. The executive producer is Manuela Saragosa. Mixing by Samantha Giovinco and Breen Turner and original scoring by Metaphor Music. Cheryl Brumley is the FT’s head of audio.

John Thornhill
Don’t forget, you can hear the rest of this season of Tech Tonic by subscribing to Tech Tonic wherever you listen to podcasts.