The Quantum Computing Question

Despite all the talk about universal quantum computers being years away, there was keen interest at this year’s Sibos conference in Sydney, which featured no fewer than eight sessions on the topic, including a curated quantum computing networking session. 

I always try to sit at the front of the room, so it came as a shock to me when, at panel after panel, all the seats were already taken, leaving me standing with several dozen other attendees in the back.

The sessions broke down quantum computing, covering quantum bits, more commonly called qubits; why quantum computing matters; what we know about it; and what we should be thinking about as we head into the future. 

David Reilly, principal research and director at Microsoft Quantum in Sydney, told the audience that building the hardware, the machine, is a tremendous challenge. “It’s not sufficient to write the equation to build the prototype. We need a much more significant approach,” he said, explaining why Microsoft is working on building topologically protected qubits. 

He continued: “Thomas Edison did not invent the electric light, but he recognized that the lightbulb needed to last long enough and he used fundamental physics and basic chemistry to do that. We are re-engineering the qubits so they can go the distance. They don’t have qubits long enough to last in the new world,” he said. 

Anna Phan, research scientist at IBM, said that while it is still early days, using a quantum approach could be used to greatly enhance computing of large mathematical problems. “Risk analysis, personal finance planning, risk or derivatives pricing—depending on the instrument of choice—needs millions of samples to get the required answer. You’d need to run risk analysis overnight or all day long using clusters of classical computers,” she said. Using a quantum approach could get the quadratic speed up to the point needed to solve large computational problems. “It’s early days, but hopefully one day there will be tools to satisfy the variety, volume, and complexity,” that financial institutions, and others, require, she continued. 

Addressing the point that quantum computers at the moment are not capable of being in a quantum state for long—the longest recorded is 90 microseconds—Alejandro Perdomo-Ortiz, senior research scientist at Rigetti Computing, noted that the device has to be kept at -273 degrees Celsius. “It is a remarkable piece of engineering. It’s important to establish the connection to classical processors. We need to control that device, so what about a hybrid approach? This is what we call the quantum-classical hybrid interface,” he said.

Rigetti is giving out a $1 million prize to the first team that is able to prove quantum advantage on its hybrid platform. Quantum advantage—a term often used by the likes of Google and IBM—is a state where a quantum system is able to solve a problem that classical computers aren’t able to. 

But what if we had perfect quantum computing devices, and as many qubits as possible? What could be accomplished? Stacey Jeffery, senior researcher at CWI, said the real potential of quantum computers is still unknown. “One real potential is machine learning and optimization problems. There are quantum techniques but we’re not sure how applicable they will be in what we do,” she said.

While there is still much up in the air about quantum computing, interest in understanding its inner workings and possibilities of what it can achieve is gradually growing, and at a much faster pace than I had thought it would. 

While there are early movers in the space—such as JP Morgan, Barclays, and Australia’s CBA—just six months ago, quantum was a foreign topic for many financial firms. But if the interest at Sibos is any indication, a tipping point might come sooner than later.