Microsoft, Majorana and quantum computing alarmism

(image:Microsoft supplied)

Last month Microsoft announced a new quantum computing chip called Majorana. Everyone sat up and blinked twice because, well, Microsoft. Also, because quantum computing has long been one of those science-fictiony technologies forever a decade or so away. Is Microsoft’s new chip a game changer? Yes. Wait, no. Maybe. Depends on who you ask. We will come back to this..

A trope bandied about the Internet suggests that the arrival of quantum computing is going to upend all manner of certainties. Again, the answer is yes and no. Quantum computing is informed by the science of particle physics and it is not really necessary to know much about it beyond the fact that it will be millions, even billions, possibly trillions, of times faster than our fastest computers, at least for certain classes of problems. How? Because what happens at tiny scales inside the atom is very odd indeed, and particle physicists and computer engineers are learning to exploit that strangeness to build these crazy fast machines.

I was recently asked whether the sudden arrival of a quantum computer would enable the bad guys to hack the Bitcoin network. My answer was that, if this were to happen tomorrow, Bitcoin would be the least of our problems. All emails, all of Dropbox, Drive and Onedrive, all credit card info, all Whatsapp and Telegram messages, all passwords, all military and financial servers would be defenceless. The entire global economic system would instantly shudder to a halt. No one will care about Bitcoin.

Cryptography is the science of keeping secrets and it has been with us for (at least) a few thousand years, ever since writing was invented. The principle is simple. You write down a secret, you scramble it and, when it is needed, you unscramble it again. It worked as well for Roman generals sending tactical messages onto the battlefield as for U-boat commanders receiving and decoding scrambled attack plans sent from shore, and then it worked again for the mathematically wondrous invention of asymmetric cryptography in the 1970s which now protects the entire digital universe.

In every case there is a ‘key’, something akin to a password, which unlocks the scramble and returns it to ‘cleartext’ (the original numbers or text or message). The current crop of algorithms which do this (to protect your browser, your crypto, your bank password, your email etc) have been engineered so that the world’s fastest computers would take thousands of years (or more) to try out every key combination before finding the right one.

Now we are being told that new machines will be millions of times faster than the fastest supercomputers currently available, and they are imminent. Why isn’t everyone screaming in panic and withdrawing their cash from banks and reverting to writing letters by hand and communicating with tin cans and wire?

There are several reasons not to panic.

The first is that the good guys have long been preparing for quantum computing - the white hats, the mathematicians and techies who (like most people) reside within a world of governance, ethics and responsibility. It has always been this way - we have always had the bad guys, black hatters and evil hackers trying to game the system, but then we have also had the cryptographers trying to stay one step ahead of the forces of darkness. The good guys have an excellent track record to date.

This remains true in the world of quantum computing. There is a large research industry pursuing quantum resistant computing. After 4 years of research and development, the US-based National Institute of Standards and Technology (NIST) has just released the final four candidates for quantum cryptographic standardisation. They are, like most successful technologies, open-source, with proposals and techniques and algorithms required to endure the stringency of exhaustive testing as well as public scrutiny. These proposals are not just ‘harder’ algorithms designed to tax even the speed of the new machines, they are instead enveloped in science and mathematics intended to make the keys uncrackable at any speed.

Which all begs the question as to when quantum computing will actually arrive. The CEO of Nvidia recently managed to crash a part of the stock market when he announced in January that quantum computing was still ten years away. (He later retracted that prediction.) In any event, quantum computing will not simply ‘arrive’. It is making its appearance gradually, starting with simple systems basically operating on a few ‘bits’ (called ‘qubits’) which have only limited bespoke applications. As the chips improve, so will the applications broaden, but they will still be restricted to quantum-solvable problems. These are generally extracted from the dark theoretical caverns of maths and science, or they are problems involving big numbers, complexity and chaos, like weather prediction. Quantum computing offers little advantage when it comes to most of the everyday computer applications we use.

There are other issues. Quantum bits decay in microseconds, so there is a limited time window to resolve a problem. Results are also error prone, requiring complex error mitigation technologies. The computer kits have to be supercooled to near absolute zero, so they won’t be on your desktop any time soon. And scaling up to higher qubit systems is not linear - the complexity grows exponentially.

Still, the quantum computing sector is well on its way and the smart money is pouring in. And so we loop back to the Microsoft announcement. How big a deal was it? Microsoft claimed to have solved several problems, including the stability and error gremlins, by using an exotic particle named after the physicist who theorised it, Italian Ettore Majorana who died (actually, disappeared) in 1938. The details of why his particle is better than the ones previously tried is buried deep in the scary weeds of physics, a place for braver commentators than I.

Some particle physicists have been a bit rolly-eyed about the announcement, making snarky comments about data and proof and repeatability and the like. Some are not convinced that what the researchers have observed is indeed the consequence of a Majorana particle. So whether Microsoft has stolen a lead on the industry is a little difficult to adjudicate at this point.

But we can say this - quantum computing is here and improving steadily. It will help solve many important human problems which require its dizzying speed to analyse and solve. It will not shut down the Internet or anything else. Those who are worrying about it can stand down.

Steven Boykey Sidley is a professor of practice at JBS at the University of Johannesburg, columnist-at-large for Daily Maverick, and partner at Bridge Capital. His new book It’s Mine: How the Crypto Industry Is Redefining Ownership is published by Maverick451 in South Africa and Legend Times Group in Europe/UK. Copy edit by Bryony Mortimer