Computers are about to hit a wall. For decades, mankind has advanced modern computing technology by packing ever more complexity — and thus computational power — into ever smaller chips and devices. The hulking machines that took up multiple rooms in the 1950s and '60s were much less powerful than the original Apple Macintoshes in the '80s, which in turn were far less powerful than today's iPhones. But that astounding run may be finally hitting its physical limits, as computing's building blocks struggle to shrink further.
So to keep its historical pace of advancement, technology may need a radical overhaul. Enter quantum computing.
By applying the behavior of subatomic particles, i.e. quantum mechanics, to the problem of computing, we might one day create devices that can solve far more complex equations much faster than today's most advanced computers. Quantum computers could eventually be used to run artificial intelligence systems; to design new materials, chemicals, and medicines at the molecular level; to model extremely complex systems, like the Earth's climate, in far greater detail; to both run encryption and crack encryption; and even render extraordinarily realistic video games.
The Week
Escape your echo chamber. Get the facts behind the news, plus analysis from multiple perspectives.
SUBSCRIBE & SAVE
Sign up for The Week's Free Newsletters
Join more than 350,000 subscribers and keep yourself informed with a selection of The Week’s most interesting, enlightening and entertaining stories - plus daily puzzles.
Join more than 350,000 subscribers and keep yourself informed with a selection of The Week’s most interesting, enlightening and entertaining stories - plus daily puzzles.
Latest Videos FromThe Week
But how do we get there?
As I mentioned, when computers were first developed in the mid-century, they were extremely primitive compared to what they've become today. But they were also extraordinarily expensive to build and maintain. That limited their use to powerful corporations and to governments. Right now, quantum computing is essentially at that same point.
Major tech companies are racing to build the biggest and baddest quantum computer they can. Venture capitalists pumped $250 million into the growing field just in 2017. World governments are also getting in on the act. In 2016, the European Union announced a $1.1 billion research project in quantum computing. Russia has the Russian Quantum Center. And the big dog is China, which will reportedly drop $11 to 15 billion on a massive quantum computing laboratory and other efforts. But it will probably be decades before quantum computers start showing up en masse in the homes of average consumers, if they ever do.
The basic problem is simple: The technology behind quantum computing is advanced, costly, and difficult to pull off.
Traditional digital computing is based on bits, the simplest unit of information a computer can store and manipulate. As 1999's The Matrix helpfully illustrated with vast arrays of flowing numerals, bits can be set in one of two ways: "1" or "0." It's basically an on/off switch, physically performed on a computer by what's known as a transistor. By packing ever more transistors (i.e. bits) onto a processor, computers become capable of solving enormously complex equations. Today's most advanced modern processors literally boast billions of transistors.
Instead of using transistors to function as bits, quantum computing uses subatomic particles. And instead of calling them "bits," people in the quantum computing field call them "qubits."
The advantage of this switch is that it's capable of far greater complexity. Subatomic particles can be in multiple states (such as a wave or particle) at once; they're a switch than can be "on" and "off" at the same time. Quantum physicists call this a "superposition." Subatomic particles can also affect each other's states instantly, a phenomenon known as "entanglement." This is very complicated stuff, but the upshot is that qubits can store much more information and perform calculations much faster than the same number of traditional bits. And each additional qubit added to a quantum computer multiplies its power more than each new bit added to a traditional computer.
IBM knocked out a 5-qubit computer in 2016, then 20-qubit and 50-qubit computers in 2017. That 50-qubit processor put IBM ahead of Intel, but then Google stepped up this past March with a 72-qubit processor. Besides the major players, there are also startups looking to get into quantum computing; one of them is apparently constructing a 128-qubit computer.
What makes these computers especially tricky to build is that other sources of energy can easily interfere with subatomic particles. For the computers to work, the superposition and entanglement of every qubit has to be tightly controlled. To keep the information stored on them stable, and to allow them to perform their calculations, qubits have to be kept in vacuum-sealed environments and at extraordinarily low temperatures, or otherwise protected from outside interference. IBM's quantum computers, for example, have to be stored in a giant fridge near absolute zero. Right now, even quantum computers with very few qubits require a lot of technological and engineering prowess.
There are different approaches to making them more resilient. D-Wave systems — which NASA and Google are working on — have reached 1,000 qubits by ignoring the problems posed by entanglement, which limits their utility to certain tasks, like finding the most efficient route between two points.
Quantum computers that are versatile and not limited are the ultimate goal. But they're still stuck around 20 qubits. And right now experts think at least 50 qubits are what's needed to achieve "quantum supremacy," where a quantum computer can reliably outmatch even the most advanced traditional computer.
Still, given how easily subatomic interference can foul up calculations, quantity of qubits is not the same thing as quality. And the field is young enough that there aren't agreed upon quality metrics yet. Both IBM and Google recently claimed they either briefly achieved quantum supremacy (before losing control of the qubits) or were on the cusp of doing so. But right now it's difficult to distinguish the achievements from marketing.
All of this gets back to why governments are probably going to play such a critical role in the future of quantum computing. Their ability to martial resources will be crucial to bringing the technology to maturity.
Which raises the question: Where is the United States government in all this?
Way behind China and Europe and the others, it turns out. The U.S. government has only allocated about $250 million — between the Energy Department and the National Science Foundation — into quantum research so far. That's led to calls for America to get its act together and start leading an international quantum research effort on par with something like the Manhattan Project. The White House is trying to put together a coordinated quantum computing research and development strategy. And there's a bill in Congress that would dedicate $1 billion to the effort.
The uncomfortable part is that quantum computing has major national security implications: It could be used to design new radar, stealth, and weapons systems. Perhaps most ominously, quantum computing promises to create encryption of stunning sophistication — and to swiftly crack any encryption not built on quantum approaches. In fact, China has reportedly constructed a communication network from Shanghai to Beijing that's protected by quantum technology and launched a communications satellite that uses it as well. And Chinese officials have made no bones about using their own research to aid their armed forces.
Technological breakthroughs by any country really ought to be seen as a win for all countries; once achieved, the technology can be shared through trade and commerce. But it's also true that countries with a technological edge tend to keep that edge for some time.
So for a country that's spending over $700 billion on the military next year alone, it's a wonder a few billion can't be found to fund the computers of the future.
