7 ways to win the quantum race
- Superconducting uses an electrical current, flowing through special semiconductor chips cooled to near absolute zero, to produce computational “qubits.” Google, IBM, and Intel are pursuing this approach, which has so far been the front-runner.
- Ion trap relies on charged atoms that are manipulated by lasers in a vacuum, which helps to reduce noisy interference that can contribute to errors. Industrial giant Honeywell is betting on this technique. So is IonQ, a startup with backing from Alphabet.
- Neutral Atom Similar to the ion-trap method, except it uses, you guessed it, neutral atoms. Physicist Mikhail Lukin’s lab at Harvard is a pioneer.
- Annealing designed to find the lowest-energy (and therefore speediest) solutions to math problems. Canadian firm D-Wave has sold multimillion-dollar machines based on the idea to Google and NASA. They’re fast, but skeptics question whether they qualify as “quantum.”
- Silicon spin uses single electrons trapped in transistors. Intel is hedging its bets between the more mature superconducting qubits and this younger, equally semiconductor-friendly method.
- Topological uses exotic, highly stable quasi-particles called “anyons.” Microsoft deems this unproven moonshot as the best candidate in the long run, though the company has yet to produce a single one.
- Photonics uses light particles sent through special silicon chips. The particles interact with one another very little (good), but can scatter and disappear (bad). Three-year-old stealth startup Psi Quantum is tinkering away on this idea.
Business Bets on a Quantum Leap
Quantum computing could help companies address problems as huge as supply chains and climate change. Here’s how IBM, Google, Microsoft, and others are racing to bring the tech from theory to practice.
by Robert Hackett
https://fortune.com/longform/business-quantum-computing/
It was a marvel of engineering, a harbinger of a future of unimaginable computational power.
It also bore a striking resemblance to a garbage can.
Q System One was a quantum computer. The machine was the culmination of a year—or decades, depending on how one measures—of labor and ingenuity from IBM scientists. The researchers had assembled this stalactite of nested canisters in the recesses of the company’s neo-futuristic research center in Yorktown Heights, N.Y. The white, refrigerated contraption dangled from a nine-foot, cubic, aluminum and steel frame. In the innermost chamber: a special processor whose progeny could help solve some of the world’s most intractable science and business problems. This particular generation featured the firepower of 20 quantum bits, or “qubits,” the powerful data units upon which these dream machines operate.
The machine was incredibly impressive, in theory; the qubits were unusually high-quality, and its error rates were relatively low—crucial advantages in the quest to make a quantum computer viable for real-life problem-solving. Granted, the thing was a little underwhelming in person, shielded in that drab receptacle. (At one meeting last November, IBM CEO Ginni Rometty remarked that it looked like a trash can.) But the scientists had a plan to get it ready for its close-up. IBM had hired a boutique London designer to shield the Q System hardware in a shiny, black metallic shell. Already, the entire contraption had been set in an air- and temperature-controlled, borosilicate glass enclosure designed by Goppion, the Milanese firm known for making display cases for the Mona Lisa and the crown jewels at the Tower of London.
By the time IBM unveiled its creation this January, at the Consumer Electronics Show in Las Vegas—a venue normally reserved for the debuts of flashy consumer gadgetry like virtual-reality headsets and phones with foldable screens—it had a supercomputer that looked super. The press and public ate it up. A “gleaming monolith from a sci-fi blockbuster,” gushed MIT Technology Review. “It looks like a computer from the future,” effused The Verge.
“Everybody takes selfies with the quantum computer,” says Dario Gil, head of IBM’s research division, who calls the technology an “object of fascination.”
Such allure, at the moment, is grounded more in hope than in results. Quantum computers can’t do much of commercial value yet; they’re still inching their way toward usefulness. The technologies that make them so potentially fast and powerful also make them, in their current iterations, unstable and error-prone compared with the so-called classical computers we rely on every day. IBM calls the Q System the “first integrated quantum computing system for commercial use,” but “use,” in this case, is highly abstract: Companies can obtain access, via the Internet, to the quantum platform at IBM’s facilities, running experiments and kicking the tires as they wait for the technology to mature.
Still, recent advances—from Silicon Valley to China, not to mention Yorktown Heights—have convinced much of the corporate world that this technology will soon move off the theoretical wish list. Companies across all industries are hoping to exploit quantum computing to surmount obstacles that have thwarted them for years. Nation-states are mobilizing, too, pouring billions of dollars into research in the hopes of gaining an edge in an area that could someday separate the world’s economic—and military—haves and have-nots. Quantum information science, which is still early in attracting private industry investment, “screams at you that it is the exact place where federal R&D dollars are best utilized,” says Michael Kratsios, President Trump’s top tech adviser and his nominee to be chief technology officer of the U.S.
The reason: The quantum computer may be our best hope of overcoming the limitations of ordinary computing. Moore’s law, the guiding principle of the tech industry, states that computing power should double roughly every two years as a result of the increase in the number of transistors a microchip can contain. But scientists are reaching the limit on how close together they can smoosh transistors on silicon chips. Everyone has been thinking, “What the heck comes after Moore’s law?” Gil says. He and many others think that quantum computing, especially in conjunction with artificial intelligence, provides an answer.
Putting quantum to work
Businesses hope to apply quantum computing to a range of complex issues; these industries are particularly eager to jump in.
Finance
Banking and investing are all about managing risk. Wall Street behemoths such as JPMorgan Chase and Goldman Sachs hope quantum computing can give them an edge in the odds, allowing them to better manage threats and opportunities related to their portfolios. Quantum computers could also help financial pros improve their Monte Carlo simulations, mathematical models designed to predict possible outcomes of complicated decision trees; they’re often used to help customers figure out how long their retirement savings will last.
Energy
Quantum computers could help the world cope with climate change, one of the world’s most complex and hard-to-predict phenomena. In January, Exxon Mobil partnered with IBM to explore applications including predictive environmental modeling and carbon-capture technology. Daimler Mercedes-Benz is using quantum computing to test new types of battery chemistry to improve electric vehicles. And the Dubai Electricity and Water Authority is working with Microsoft to optimize its energy grid management.
Medicine
One day, your health may depend on a quantum leap. Pharmaceutical giant Biogen teamed up with consultancy Accenture and startup 1QBit on a quantum computing experiment in 2017 aimed at molecular modeling, one of the more complex disciplines in medicine. The goal: finding candidate drugs to treat neurodegenerative diseases. Microsoft is collaborating with Case Western Reserve University to improve the accuracy of MRI machines, which help detect cancer, using so-called quantum-inspired algorithms.
A milestone test is not far ahead. Google believes it will reach “quantum supremacy”—a stunt-like demonstration of a machine’s superiority over a traditional computer—in the very near term. Chinese scientists say they’re on a similar timeline. Once that bar is cleared, “businesses and technologists will look at that and realize it’s not just some promising technology in the future, but something powerful working right now,” says John Martinis, who leads Google’s quantum efforts.
Even in the absence of that confirmation, there’s a land grab underway. IBM is jockeying with Google, Intel, Microsoft, and a host of other tech giants and upstarts to dominate the territory. If these companies can convince people that they have the right approach, they will win over more developers, more prospective customers, and more market share. Not coincidentally, many of these companies rent out or host software and services “in the cloud” for other computer users: A quantum breakthrough would give them another potentially profitable service to offer.
“I get a lot of questions from customers about when is quantum coming and when is this applicable to my business,” says Julie Love, Microsoft’s quantum business development leader. “Increasingly, we’re saying ‘Today.’”
* * * *
The potential is so enticing because a quantum computer is not just another ultrafast computer: It’s a new beast entirely. Instead of computing one thing after another, plodding along brute-force style as regular computers do, quantum computers could potentially consider all scenarios simultaneously, like a monk who has attained nirvana through meditation.
To understand the kinds of problems quantum computers are theoretically suited to solving, imagine standing in the Alps, looking at the mountaintops. Ask yourself: Which one has the highest peak? A simple scan of the horizon yields the answer (and centuries-old trigonometry can confirm it). Now try to imagine a universe with thousands of dimensions—or better yet, hundreds of thousands—rather than the standard three with which we’re familiar. Discovering any given minimum or maximum point in this kaleidoscopic hellscape is effectively impossible.
More companies these days find themselves in the thousand-dimensional Alps. They’re awash in data, to be sure. But even the most powerful computers can’t solve some kinds of problems, because they involve too many kinds of data—too many variables.
Consider Amazon, which seeks to ship everything to everyone as efficiently as possible. Someone trying to “optimize” that effort has to deal with countless questions involving routing and logistics, inventory, weather, traffic, local laws, and whatever else the universe throws at them. Humans and traditional computers wrestle with the chaos as best they can: A quantum computer might tame it. And boosters see even more potential for the tech in tandem with A.I.: As self-teaching machines take on more responsibilities, quantum computing could turbocharge machine-learning processes.
The possibilities are seemingly endless, and also unproven, which is why the tech lends itself to the inflationary churn of the hype machine. But that isn’t stopping companies from coding up software so they will be prepared when the real deal, a so-called universal quantum computer, comes online.
JPMorgan Chase and Goldman Sachs are exploring quantum applications to manage risk in investment portfolios. Daimler Mercedes-Benz hopes to use the technology to boost battery performance in electric vehicles. Pharmaceutical giant Biogen has run quantum-driven tests to find new candidate drugs to treat neurodegenerative diseases such as Alzheimer’s and Parkinson’s. It’s easy to see why so many companies are so invested in this burgeoning market; perhaps no other emerging technology spans so many different disciplines with so many potential applications.
“We’re getting in on the ground floor,” says Vijay Swarup, vice president of research and development at Exxon Mobil. The energy giant announced a partnership with IBM in January in Las Vegas, in tandem with the Q System’s splashy debut. Swarup’s company sees applications in making environmental predictions, optimizing energy grids, and generating breakthroughs in carbon-capture technologies. “Quantum computing can take our understanding of nature and chemistry to a granularity that has never been able to be done before because the computations are just too hard,” Swarup says.
* * * *
The idea for a quantum computer has been around since at least the ’70s. Today, the most optimistic practitioners will tell you that the obstacles are increasingly engineering-related, as scientists try to figure out how to make the machines work reliably and at scale. As Pedram Roushan, a member of Google’s quantum unit, puts it, “People are still puzzled by the principle of quantum mechanics, but they’re going to live with it and try to put it to some use.”
In 1995, Peter Shor, a mathematician then at Bell Labs in New Jersey, proved that a fully functional quantum computer could do something remarkable: It could crack RSA encryption, a popular means of securing private communications. He showed that his quantum algorithm could do in minutes what might take a regular computer the lifetime of the universe to unravel. A year later, Lov Grover, also a Bell Labs scientist, came up with a quantum algorithm that would allow people to swiftly search unstructured databases. Scientists piled into the field, and advances in hardware soon followed the breakthroughs in code.
By the mid-2000s, a team led by Robert Schoelkopf of Yale, whose lab would eventually seed the quantum field with executives and scientists, devised an approach to quantum computing upon which the tech world’s greatest hopes hang today. Schoelkopf helped pioneer a so-called superconducting qubit, which uses supercooled silicon and electrical currents to work its magic. IBM’s machines are a direct descendant of Schoelkopf’s lab. Rigetti Computing, a California startup led by Chad Rigetti, a Schoelkopf lab alumnus who formerly played a key role in the quantum computing effort at IBM, builds machines of this type, including a 128-qubit one it plans to debut later this year. Google’s and Intel’s foundations also rest on this technology.
One reason the approach is so popular is because it builds atop decades of advances in the semiconductor industry. These qubits are created inside specially designed silicon devices; they’re generated by an electrical current flowing between superconducting electrodes separated by a thin insulating barrier. (This works only in cryostatic, ultracold chambers, which helps explain why quantum computers will live for the foreseeable future in labs and data centers, not on desktops.)
When someone operating a quantum computer enters certain commands, they can link two qubits together, entwining them in a state called “entanglement.” If something happens to one entangled qubit, its mate instantaneously reacts. By stitching together networks of such qubits, a programmer can run massively parallel operations, meaning a huge number of operations at once. This is what enables quantum computing’s exponential speedups.
“Superposition,” a related concept, is the other key to quantum computing. Whereas bits, the building blocks of classical computing, are limited to representing information as “zeroes” and “ones,” qubits can assume any combination of gradations between zero and one. Think of this as the difference between a coin at rest on a table, displaying heads or tails, vs. one spinning, ballerina-like, on its edge. The result: Superposition allows qubits to store vast amounts of data compared with regular bits.
Together, superposition and entanglement give quantum computing its kick—amplified memory tackling complex problems at remarkable speeds. (The trick only works while no one is watching, a bizarre but fundamental fact of quantum science. As soon as someone observes the system, everything collapses.) The act of measurement causes a cascade of tipped-over qubits that produces a final state. If the math is right and the machine well-designed, then that system should tend toward the most probable, most optimal state—the solution.
Each qubit adds exponential power. But as the quantity of qubits grows, quality becomes a limiting factor. As with a spinning coin, even the most minor disturbances, such as heat or vibrations, can shake up the system, causing errors that manifest as wrong answers. And in today’s machinery, as the number of qubits increase, so do error rates. Indeed, some practitioners fear there may be a fundamental law, as yet unknown, prohibiting these machines from working at scale—like Jenga towers, they may be doomed to tumble when they get too high. Some skeptics, such as Gil Kalai, a professor at Hebrew University in Israel, believe that the technology will never work as hoped: “My analysis suggests that efforts to build quantum computers are going to fail,” Kalai says.
That tension explains why IBM and Google are so eager to demonstrate that they’ve fortified their qubits and lowered their error rates. It also explains why other scientists are exploring the possibility of a better way forward.
* * * *
Chris Monroe, a physics professor at the University of Maryland, remembers the cold-call email in February 2014 that changed his professional destiny. The correspondent was an investor who sought a meeting. Monroe had published a paper that month in a prominent physics journal, effectively outlining a road map for how certain devices could help quantum computing leap forward. When the visitor showed up at Monroe’s office, he brought the article with him. This wasn’t just a science paper, the man said, waving the document in the air. “This is a business plan!”
That investor was Harry Weller, a partner at the venture capital firm New Enterprise Associates (NEA), a legendarily successful early backer of shopping site Groupon and a passel of software startups. Monroe, who had been contentedly sustaining his academic research with grants from the U.S. intelligence community, wasn’t interested at first. Eventually, though, he came around and accepted Weller’s funding proposal, founding IonQ in 2015. (Weller died in 2016.)
IonQ is working on an approach to quantum computing, described in Monroe’s paper, called the “ion-trap” method. It activates the qubits in its system by manipulating ions, or charged atoms, with laser beams. In the ion-trap method, unlike with superconducting qubits, physical wires are not needed to send control signals into the machine. That means the qubits are better protected from “noise,” or disturbances that contribute to error, Monroe says. They sit suspended in a vacuum cushion, like a maglev train hovering on its tracks. GV, the venture capital arm of Google parent Alphabet, joined NEA as an IonQ investor in 2017. In May, the company added the former director of engineering of Amazon Prime as its CEO.
The ion-trap idea has some prominent converts. Honeywell, the industrial conglomerate, last year debuted an ion-trap approach that it had been working on in secret for years—a major point of validation for Monroe’s startup. Honeywell found that its expertise in areas like vacuum systems, lasers and optics, microelectronics fabrication, and other disciplines all converged in the new field. “If you put all those things together, you can build a quantum computer,” says Tony Uttley, who leads Honeywell’s 100-person quantum efforts.
The ion-trap method is only one of more than half a dozen approaches to quantum computing. It has produced promising early results. In this nascent field, of course, it’s difficult to compare the performance of one technology with another; scientists even disagree about where to begin to do so. And it’s far too early to predict which approach might become dominant.
For students of history, the transistor provides an instructive metaphor. The device, invented in 1947, went on to become the foundation for all modern computers, but few could have predicted the extent of its significance at the time. When the transistor debuted, the New York Times covered it in a very brief article tucked away on page 46. The front-running technologies in computing at that time: vacuum tubes and relay circuits. And if you had been picking winners back then, you might have ignored the transistor.
7 ways to win the quantum race
There are multiple ways that quantum computing could work. Here’s a guide to which companies are backing which tech.
Superconducting uses an electrical current, flowing through special semiconductor chips cooled to near absolute zero, to produce computational “qubits.” Google, IBM, and Intel are pursuing this approach, which has so far been the front-runner.
Ion trap relies on charged atoms that are manipulated by lasers in a vacuum, which helps to reduce noisy interference that can contribute to errors. Industrial giant Honeywell is betting on this technique. So is IonQ, a startup with backing from Alphabet.
Neutral Atom Similar to the ion-trap method, except it uses, you guessed it, neutral atoms. Physicist Mikhail Lukin’s lab at Harvard is a pioneer.
Annealing designed to find the lowest-energy (and therefore speediest) solutions to math problems. Canadian firm D-Wave has sold multimillion-dollar machines based on the idea to Google and NASA. They’re fast, but skeptics question whether they qualify as “quantum.”
Silicon spin uses single electrons trapped in transistors. Intel is hedging its bets between the more mature superconducting qubits and this younger, equally semiconductor-friendly method.
Topological uses exotic, highly stable quasi-particles called “anyons.” Microsoft deems this unproven moonshot as the best candidate in the long run, though the company has yet to produce a single one.
Photonics uses light particles sent through special silicon chips. The particles interact with one another very little (good), but can scatter and disappear (bad). Three-year-old stealth startup Psi Quantum is tinkering away on this idea.
Dave Wineland, a Nobel Prize–winning scientist who coinvented the ion-trap approach, frames the issue with a different metaphor. “It’s like starting a marathon race. Maybe ion traps are in the lead, but we can still look behind us and see the starting line.”
Growing numbers of corporations are deciding they can’t wait five or 10 years for a winner. Daimler Mercedes-Benz, which has been partnering with both IBM and Google on quantum research, is one such believer. “There are certain simulations and modelings that we cannot achieve with current computing power,” says Ben Boeser, innovation director for the company’s North American R&D unit. Daimler hopes to use quantum techniques for optimizing transportation logistics and modeling the chemistry of vehicle batteries. Such calculations remain out of reach for quantum computers today, but Boeser’s team expects the technology to get there in the coming years. “We believe if we don’t jump in as an industry giant, the technology partners may not put their emphasis on those use cases, and hence we would miss out.”
“People are ignoring these problems now because we don’t have the machines to do it,” Monroe says. Eventually, “people will start thinking about these computations more if there’s even a hint of advantage, and that’s going to snowball.”
You don’t need a Ph.D. to take these machines for a spin. Since 2016, IBM has made two quantum computers accessible to the public via a website with a graphical interface that looks like a musical score. Scientists inside and outside the corporate world are running experiments via similar portals. They’re exploring approaches to optimization problems, trying to figure out what sorts of questions they can ask and how they’ll frame those questions once the technology is further along. In three years, 120,000 people have performed more than 10 million experiments and published more than 190 research papers using IBM’s so-called quantum cloud service.
During a mid-December visit to IBM’s Yorktown Heights facility, the research center’s staff showed off a time-lapsed heat map of the world. The geography reveals who has been dabbling on the computers. Everywhere, enthusiasts are learning, coding, and experimenting. Except for an apparent anomaly: On the heat map, China remains surprisingly dark, despite its size, influence, and interest in the technology. Here be dragons…
Dario Gil, the research center’s chief, acknowledges the paucity of activity on the other side of the world. The Chinese have their own government-spearheaded initiatives, and they are not working with American corporations, at least not IBM, he says.
Gil’s remark is a reminder that the quantum competition is not merely commercial—it’s also geopolitical. The first country to build a fully functional, general-purpose quantum computer may be able to pierce the encryption that protects Internet traffic and secures all variety of data, an invaluable tool for spies. Countries at the forefront of the technology may also be able to eavesdrop-proof their communications, an obvious advantage in a geo-rivalry.
The competition heated up in 2016, when Chinese scientists blasted a satellite into low-earth orbit. Within a year, these scientists used the spacecraft, nicknamed Micius after an ancient Chinese philosopher, to successfully transmit so-called quantum entangled particles more than a thousand kilometers between the skies overhead and the Tibetan mountains on Earth. The world marveled at the feat, and spines tingled: Had America lost its lead in this contest so soon, just as it had once seemingly fallen behind the Soviet Union in the space race?
China has activated a highly secure “quantum key” communications line, between Beijing and Shanghai. Since 2013, the Chinese have published nearly 500 more papers than their American counterparts on quantum science: 2,986 vs. 2,494, by Boston Consulting Group’s count. Moreover, China’s government is said to be spending $10 billion over the next five years on a national quantum program. Anton Zeilinger, an Austrian physicist who taught Pan Jianwei, the scientist who led the Micius expedition, tells Fortune that, with respect to quantum communication, “it’s safe to say that China is ahead of the game. And not just by a small increment.”
Back in the U.S., politicians have gotten the message. At the end of 2018, just before a gridlock that resulted in the longest-ever federal government shutdown, Congress, with near unanimity, passed the National Quantum Initiative, a bill authorizing more than $1 billion to kick-start an American national quantum strategy. The initiative coordinates funding activities across major research agencies. It’s as yet undecided how the money will be spent, but the injection of federal funds is both a vote of confidence in the technology and a powerful motivator for funding-hungry research labs.
Many Americans disagree with the notion that China has the edge, given the pioneering work of U.S. corporations, universities, scientists, and startups. Kratsios, the U.S. chief technology officer designate, says that other countries are pouring tremendous sums of money into quantum science because they’re behind, “playing catch-up.” Regardless of who’s currently leading, Joe Broz, a theoretical physicist who leads the advanced technology division at SRI International, an influential laboratory group born out of Stanford University, says the act will give the U.S. the ability to nurture the nascent industry and prevent it from “escaping offshore to our detriment, where it’s only to be sold back to us.”
As with any early-stage technology that presages a revolution, there’s always a risk that the hype exceeds the hope. (See the frequent periodicity of A.I. winters, when advances in that technology have slowed drastically, dampening enthusiasm along with it.) Some scientists worry that investment will run dry once investors encounter extended timelines and delays on the product road map. “There’s a joke in quantum computing that it’s always five years away,” says Matthew Brisse, a Gartner analyst, pointing to decade-old headlines that claim a breakthrough is just around the corner.
If this time turns out to be different, it may be because so many companies are putting their shoulders to the wheel. “We’re on the heels of a new industry forming,” says SRI’s Broz. “I can imagine how people felt in the ’50s, ’60s, and ’70s with the semiconductor industry emerging.”
As for timelines, Jim Clarke, the head of quantum hardware at Intel, draws an analogy to both the mission to put a man on the Moon and the development of modern electronic computers. Sputnik flew in 1957; Neil Armstrong touched down on the Moon in 1969. The first transistor came about in 1947; the first integrated circuit arrived in 1958. Such transformational leaps typically take a little over a decade, and the quantum computer will be no different, Clarke forecasts.
“We’re not trying to meet some short-term, flashy goal, but we’re trying to build that rocket ship to the Moon,” he says. Nobody can quite agree on when the industry will see liftoff, but this could be the year scientists start the countdown.