“I think I can safely say that nobody understands quantum mechanics,” said Nobel Prize-winning theoretical physicist Richard Fenyman. His sentiments were echoed by Bill Gates, who said recently that Microsoft’s quantum-computing project is “the one part of Microsoft where they put up slides that I truly don’t understand.” Even Albert Einstein couldn’t wrap his head around it, so to speak, declaring that the thinking behind quantum mechanics was fundamentally flawed.
And yet… Scientists have since proved the theory repeatedly and conclusively. More to the point, computers today are being built upon the principles of quantum mechanics – they actually exist. And wouldn’t you know it, Google (a division now of Alphabet, Inc.) is taking a lead role. (So is China.)
The oversimplified guiding principle here is that in the quantum world – the world at the atomic level – a strange phenomenon exists known as ‘superposition’ which states that a single atom can be in two locations at the same time. In our ‘real’ world, that’s simply impossible, and so we can excuse Fenynman, Gates, Einstein and ourselves for not ‘getting it.’
But as Vijay Pande, a partner at Andreessen Horowitz, the Silicon Valley venture firm says, “If this works, it will change the world and how things are done.” He’s not kidding.
How these things work is less important to most of us than what their working may portend for the future. At the ‘how’ level, quantum computing involves using qubits rather than the traditional computer’s bits. In bits, everything is either a one or zero. In qubits, they can be a one and a zero at the same time. This allows qubits to process a lot more information than bits, which are set in a specific state – exponentially so when they are combined. Thus, while 1 qubit can equal 2 bits, and 2 can equal 4, by the time you get to 10 qubits you have the ‘equivalent’ (loosely speaking) of 1,024 bits. And at 20 qubits, you have over 1 million equivalent bits. You get the idea.
The practical application of this compounding expansion of computing technology becomes very relevant when you start looking at really deep and complicated problems, for example unbreakable encryption, or simply creating algorithms to calculate the fastest routes to the airport with minimum traffic. A classical computer would take 45 minutes to take the data of 10,000 taxis and perform that task; an experiment with a quantum computer from the Canadian firm D-Wave (whose quantum PC is pictured above) did it in less than a second.
Because large-scale encryption can be enabled (reverse-factoring prime numbers is a common encryption technique) as well as unbound (or cracked), quantum computing power has attracted the attention of the NSA, where code-breaking quantum computers could be devastating to the national security. NSA employees and vendors have already been put on alert that they will soon need to overhaul their encryption techniques. Of course, the NSA (like Google) is building its own quantum computer.
The potential exists to upend entire industries, which of course is why Google is employing its quantum computing powers in the realm of AI. Word is they already have a 22 qubit chip, frozen inside ‘cryostats’ in Santa Barbara, and that they plan to use their complex (and expensive) setup to deliver quantum computing via the cloud, possibly charging by the second according to reporters at The Wall Street Journal.
There are still some very real hurdles to overcome, from error-checking to the expensive containers for the chips used to power then, but if computers have taught us anything the past 50 years, it is that the future will always be faster, and hence more powerful – and the sky, or perhaps more to the point, the atom – is truly the limit.