New Tiny Computers Could Have A Huge Impact
MARY LOUISE KELLY, HOST:
It feels like computers are getting smaller all the time. What used to fill a room now fits in your pocket. Now some companies are betting big on new ones that run at the atomic level. NPR's Geoff Brumfiel has more on how these tiny machines could have a huge impact on our lives.
GEOFF BRUMFIEL, BYLINE: Inside a bland-looking suburban office building near Washington, D.C., Christopher Monroe is working on the future.
CHRISTOPHER MONROE: We actually only recently expanded, taking over the entire building. We only had a third of it before.
BRUMFIEL: Monroe is chief scientist at a company called IonQ. IonQ is trying to develop an entirely new kind of computer known as a quantum computer. In a glass display case in the lobby is one of the early prototypes.
MONROE: So if you look closely in that little circle there, you might be able to see a chip, a silicon chip.
BRUMFIEL: The microchip looks a lot like those in other computers, but the thing is that's not what makes this quantum computer tick. The machine does calculations using individual atoms.
MONROE: Atoms are basically glued to that surface. They're not glued. They're just levitating.
BRUMFIEL: Above the microchip and then poked and prodded with lasers. Think of it as an atomic abacus. And the hope is that this kind of computer will be able to provide answers to really complex questions.
MONROE: We're building something that's going to hit everything - oil and gas, energy, Big Pharma, financials, logistics. Almost any company that has a hard problem, which is everybody, they're going to be able to use quantum computers for it.
BRUMFIEL: Quantum computers are machines that calculate using the rules of a theory called quantum mechanics. These rules govern very small things like individual atoms. Marissa Giustina works on quantum computers for Google.
MARISSA GIUSTINA: Because our intuition doesn't operate according to quantum mechanics, it's kind of difficult to get a feeling for how the rules are different. But there are things we gain and things we lose, I would say.
BRUMFIEL: The gains these computers are capable of are incredible. In 2019, Google's machine solved a problem in three minutes that would take a normal supercomputer up to 10,000 years. It sounds impressive, but the problem was an abstract math equation - nothing particularly useful. And there are a lot of other problems that normal computers are way better at solving than quantum machines. Simply counting numbers, for example - one, two, three - quantum computers are terrible at it.
GIUSTINA: It's just hard. Our could count to, I think, four a couple of years ago. So (laughter)...
RAJIBUL ISLAM: In the future, I don't think that everybody will stop using classical machines and go to quantum. That's not going to happen.
BRUMFIEL: Rajibul Islam is a researcher at the University of Waterloo in Canada. He says that normal computers aren't going to disappear, but there is evidence that quantum computers might be able to do some things that are impossible right now - break some kinds of encryption, for example, discover new materials in medicines or figure out the best way to send out packages from a central warehouse. All these difficult problems might be solvable if scientists could just build a better quantum computer.
ISLAM: We know where quantum computers will make a difference. It's a matter of high-quality engineering and then making the systems better and better so that we get there.
BRUMFIEL: At the moment, nobody's quite sure which application will be the killer app. But Google, IBM and other big companies are all investing heavily to make more powerful, more reliable machines. And so is IonQ. The startup has raised about 80 million in capital, and Monroe and his team are working hard on their latest quantum machines. He takes me to see them in a large room at the back of the building.
MONROE: These are a couple of the next-generation systems that are being built right now, actually.
BRUMFIEL: Even though IonQ's computers haven't solved any impossible problems just yet, Monroe says he's not losing any sleep.
MONROE: There's so many very hard problems out there that regular computers can't touch because they're too hard. If you're going to have any hope with those problems, it has to be quantum computing.
BRUMFIEL: Monroe and others believe the future is quantum. They're just not exactly sure when that future will come. Geoff Brumfiel, NPR News. Transcript provided by NPR, Copyright NPR.