The Current - Meet Willow, Google’s new quantum computing chip

Episode Date: December 13, 2024

Google promises its new “Willow” chip — unveiled this week as the latest update in the field of quantum computers — might someday revolutionize everything from drug discoveries to artificial intelligence. A technology reporter explains what quantum computing is, and why the arms race is on, even though practical uses are still years away.

In 2017, it felt like drugs were everywhere in the news, so I started a podcast called On Drugs. We covered a lot of ground over two seasons, but there are still so many more stories to tell. I'm Jeff Turner, and I'm back with Season 3 of On Drugs. And this time, it's going to get personal. I don't know who Sober Jeff is. I don't even know if I like that guy.

On Drugs is available now wherever you get your podcasts. This is a CBC Podcast. Hello, I'm Matt Galloway, and this is The Current Podcast. Imagine a machine so powerful that it can solve an extremely complicated math problem in just a few minutes. Sounds pretty impressive. Now, what if I told you that to perform that same task, it would take one of the world's most powerful supercomputers

longer than the amount of time the universe has existed? This week, Google announced that it got a machine to perform that same math calculation in under five minutes. The machine is a quantum computer with a chip they've named Willow. This is the kind of thing people in the tech industry have been striving toward for decades. And it also involves another extreme,

very, very low temperatures. Have a listen to one of Google's engineers, Ashley Huff, explaining in a corporate video how cold they need to keep this computer. The farther down you go, the colder you get. We start at room temperature, 50 Kelvin, 10 Kelvin, 3 Kelvin, which is outer space now for a reference, 1 Kelvin, 0.1 Kelvin, 0.01 Kelvin. And this here below the plate is where we actually mount the quantum processor. Literally the coldest places in the universe.

It's minus 273 degrees Celsius. It's not just cool, though. Quantum computing is now a step closer to making a real difference in our lives, for better or for worse. Cade Metz is a technology reporter for The New York Times. Cade, good morning. Good morning. Before we talk about what this thing can or can't do,

can you just explain what quantum computing is? Wow, now that is a difficult question. It's such a fascinating thing, but it's a hard thing to grasp. A quantum computer behaves according to the laws of what scientists call quantum mechanics. And it's the physics of very tiny things like an atom or very, very cold things like the system at Google that you just described. These quantum systems behave in ways that things in our daily lives, the world of classical physics, don't behave. The fundamental idea is that whereas a classical computer, the kind of computer you may be sitting in front of right now, it stores information in what are called bits. This is the fundamental building block of a computer.

Each bit can store either a one or a zero. And as you string those bits together, you can store more numbers, more ones and zeros. And you can perform calculations on those numbers. It's a lot like what you would do in high school on a piece of paper. It's just done at a larger scale and faster on your laptop. With a quantum computer, you use those quantum systems I talked about,

those very tiny things or those very cold things to store your information. The trick is a quantum A quantum system can store a one and a zero at the same time. It can hold two numbers simultaneously, physically. That's not something that you or I can relate to here in our everyday world. Things are either one thing or another. They can't be two things at the same time. Things are either one thing or another. They can't be two things at the same time. But believe it or not, in these systems, something like Google's chip, machine becomes exponentially more powerful than anything we could build here in the classical world. So they did this math calculation in some ways to prove that point with this chip that is kept at this insanely low temperature.

Why is this such a big deal? You phrased it well. It's to prove the point. It's to show that after decades of research, quantum machines can be that much more powerful than a classical computer, right? They're showing that there is a calculation that a quantum machine can perform in five minutes that a classical supercomputer, one of the most powerful supercomputers on Earth today, couldn't perform in 10 septillion years, which sounds like something out of a sci-fi movie. But they phrase it that way to make their point, right? This calculation is not something practical. It's not something that's going to change your life or anyone else's life today. But it shows what's possible in the future.

There's still a hurdle to get over, a large hurdle, but we can talk about that as well. What are the hurdles? Because people have been thinking about this for a long time. I don't know whether they had faith that it would actually work, but people have been striving toward this, as I said, for a long time. What are the hurdles to get it to a practical use? Well, this is another mind-bending concept that can be hard to wrap your head around. But I talked about that phenomenon where in the quantum world, something can be two things at the same time. The trouble is, if something like you or I, a classical piece of physics, tries to look at that system and read both things, it decoheres, as scientists say.

It collapses into an ordinary classical system that can only hold one thing. So just by trying to read your quantum computer, you destroy it. It's no longer a quantum computer. And that seems impossible to get around, if you think about it. Scientists for decades have been trying to get around that problem. And what they aim to do is put lots of these qubits together and find ways of reading that information, maybe read it out of some but not others, and take the calculation of the whole system

as opposed to just one qubit. Essentially what they're trying to do is they're trying to reduce the errors that come when you use a quantum computer. When you look at some of that data, it creates errors. It creates mistakes that you can't get over. But what Google also showed recently with this quantum machine is that it can reduce the errors in a significant way. They cross what they call the error correction threshold. And all that means is that they have shown that as they put more and more of these qubits together, that the number of errors reduces exponentially. So the larger the quantum machine, the fewer the errors. What that means in the long run is that we're on a path

towards a system where the errors are no longer a huge problem and you can do things that are practical. You spoke with a theoretical physicist for your piece in the Times who said, we will eventually see the impact of quantum computing in our everyday lives. How will we see that? What is the promise here? The promises are big, right? The promise is that eventually this will help us do drug discovery. So develop new medicines and vaccines, that it will help us develop artificial intelligence, something that's in the news every day, and improve these systems that in automated ways help us do things in our everyday lives. But there's also a flip side.

There's a danger to this. Scientists have long worried that if you have a quantum computer, a viable quantum computer, it can break all the encryption on Earth that protects the world's secrets, that provides security for governments. That moment is something that scientists have long feared and that could be brought about by a quantum computer given that how much of an arms race is there underway right now to try to not just develop this but to your point reduce the errors such that it is it is of practical use it's an enormous arms race it's here in the u.., it's not just companies like Google that are doing this. Rivals like IBM, Microsoft, and Intel are building similar machines. This is a big area of interest in Europe, and it's a big area of interest in China, which by some calculations, the Chinese government's own calculations, has contributed over $15 billion to this type of technology.

All of these organizations are working towards building quantum computers that could potentially break encryption, but also new quantum systems that would provide new methods of encryption capable of standing up to a quantum computer. You write a lot about AI and you mentioned this already that it could help advance perhaps artificial intelligence, but do you see a similarity comparable kind of narrative here in part because we talk a lot about AI, we don't talk nearly as much about quantum computing. But do you see a similarity in terms of how, if scientists can smooth out those bumps in the road, that could change our lives? It could. And change works in two ways, though. It can change our lives for the better or for worse.

And that's why they're similar. You do hear more about AI nowadays because it is more real, so to speak. And we've used chatbots now for a couple of years. The average person has. And we can see a little bit of the power and the potential of this technology. A quantum computer is still very much in the future. Let's make that clear. Getting to the point where it can do something

that's going to change our lives, maybe years away, it could be a decade away, maybe more than that. So we'll see how this plays out. Scientists are often overly optimistic about these things, whether it's AI or quantum computing. Cade, thank you very much. Glad to be here.

Cade Metz is a technology reporter for the New York Times and the author of Genius Makers, The Mavericks Who Brought AI to Google, Facebook, and the World. He is in Berkeley, California. For more CBC podcasts, go to cbc.ca slash podcasts.