Quantum computing: a problem for another generation?

In the final act of the science-fiction classic Back to the Future, Marty McFly, played by Michael J. Fox, performs Chuck Berry’s 1958 rock song Johnny B. Goode at his parents’ high school dance—in 1955. As the song ends, McFly includes more anachronistic rock star moves, including Berry’s duckwalk, and playing the guitar behind his back a la Jimi Hendrix, before finally windmill strumming the guitar and kicking the speaker like Pete Townshend. As McFly looks out over the stunned and silent audience, he says, “I guess you guys aren’t ready for that yet—but your kids are gonna love it.”

Understanding that emerging technologies may have future commercial advantages but being unable to realize them today is the crux of the quantum computing problem inside banks. Earlier this year, UBS’s former chief data officer and head of the bank’s innovation lab, Lee Fulmer, told attendees at a conference hosted by The Economist that the bank had abandoned a multi-year effort to utilize quantum computing for trading after concluding it currently offered no advantage over other technologies.

“In investment banking, you live and die by microseconds,” he said. “The end result of all of that effort was that we found we weren’t getting a substantive uplift.” UBS had concluded that the technology did not offer an immediate way of either speeding up existing machine-learning models or giving the bank an advantage over its competitors.

The decision made by a large bank like UBS raises questions about the potential merits of pursuing quantum computing projects in the banking industry. With quantum computers not expected to be commercially viable for at least—by some estimates—another 10 to 15 years, there is growing debate on whether banks should reconsider their current investments and strategies related to quantum technology.

The progress of quantum computing has been simultaneously impressive, but also underwhelming from the perspective of listening to what the quantum computing industry said would happen in the past

Fabio Sanches, Federal Reserve

A quant at a tier-one US bank that has been experimenting with quantum computing says that without the correct infrastructure behind it, investing in quantum computing is “not meaningful”.

“Quantum computing is amazing, but it cannot operate at high speed because we don’t have the right infrastructure for it,” he says. “It’s like if you have a Ferrari in London—it’s a great car, but you can’t speed because London is full of potholes and your car will break.”

In contrast with classical computing, quantum computing, in a nutshell, differs at the bit level. In a classical computer, bits store either a one or a zero, but a quantum bit, or ‘qubit’, can exist in both states simultaneously. This means quantum computers can handle far more calculations and computations than classical computers, but at present they are far more prone to error.

This is due to numerous specific parameters that qubits demand over classical bits. For example, if qubits get too warm, due to stray waves or manufacturing mistakes, the number of errors increases.

For Franco Severini, chief technology officer at multinational Japanese IT services provider Fujitsu, quantum computing’s development is similar to the early production of microchips.

“From the creation of the first transistor to its integration onto a wafer at a reasonable size took about 10 years, and then it was another 10-15 years or so before the first commercial microchip became available, with which you could do something,” Severini says. “We’re looking at the same kind of timescale. Just because something is far out, doesn’t mean it’s not worth pursuing.”

Severini says something any large organization needs to account for is the timeline for when quantum computing capabilities will become utilizable to provide real business advantage. He says that point on the timeline has not yet been reached, and isn’t very close, so businesses need to decide how many of their resources they will allocate towards keeping an active presence in quantum computing until then.

“That might be just as simple as maintaining a small team within your organization of just a few individuals who are quantum aware, who can track movements in the industry,” he says. “They can be constantly learning and iterating and publishing internally in the organization, so that when it gets to the point when you can act, you’re able to leap on it and get that advantage.”

Severini posits that for a technology as potentially game-changing as quantum computing, 10 years is not a particularly long time. In 10 years’ time, he says, companies that are acting now to pursue quantum computing will be at the forefront of the industry and be best positioned to take advantage of it.

Fujitsu is one of a handful of vendors that allow financial services firms to experiment with quantum solutions for research by using its own quantum computer on a Software-as-a-Service model. Earlier this year, Fujitsu made available its hybrid quantum computing platform, combining the power of its 64-qubit quantum computer with a 40-qubit quantum simulator, to clients through a unified API layer. Severini says that this model allows banks to circumvent the high costs that come with developing unique proprietary quantum computers.

“Any long-term adoption cost of buying your own quantum computer would be high, of course, because you’re talking about very rare things that are expensive and difficult to make and have a lot of environmental and physical demands around their construction. But that’s not an option that a bank would take in a real-world business scenario. They would use a provider’s quantum services,” he says.

Money-spinning in the quantum realm

One of the problems with quantum computing is that the subject is hard to grasp, and its implications are harder to parse. In other words, the barrier to entry is high. And this creates challenges in communicating the potential benefits—and pitfalls—of maintaining a quantum computing team, especially when such a team’s output is not immediately obvious as a business advantage, says Taha Jaffer, head of wholesale banking and global treasury AI at Scotiabank.

“When you say something like quantum, not only do you instantly think ‘science experiment’, you have trouble getting to a place where you can figure out where the connection to the business is.”

After working in the capital markets and at a variety of hedge funds over more than 20 years, Jaffer is a big believer in the speed increase that quantum technology has the potential to bring to financial services.

It’s like if you have a Ferrari in London, it’s a great car but you can’t speed because London is full of potholes and your car will break

Quant at tier-one US bank

“Even if you don’t get to quantum advantage, there is still a very strong part of the community that feels like you will still be better [on quantum] than you could do on a regular CPU—even if it’s not five orders of magnitude faster.”

Jaffer says Scotiabank is achieving meaningful returns out of its investments in quantum computing and has a product currently in the proof-of-value stage. He believes that quantum computing is going to experience a “series of ChatGPT-like moments”—referencing OpenAI’s flagship generative AI application that kickstarted much excitement around the potential impacts of the emerging technology.

“In most of these major banks, they’re spending a lot of time not putting any more money into science experiments, which is the learning that came out of the AI buzz, right? There was a lot of money spent, and a lot of money wasn’t made,” Jaffer says. “Banks have been making money for such a long time that if there is something disruptive, they’re probably just going to go and buy it later.”

How soon is now?

Quantum computing is talked about as a definite feature of the future, and there is an expectation that it will revolutionize many industries, including financial services. But UBS’s decision to drop its quantum effort was not met with calls of shortsightedness nor derision from those within the quantum field.

Fabio Sanches, director of quantum computing at the US Federal Reserve, thinks that it’s a reasonable decision. He says it’s still early days for quantum computing in financial services, and despite some experiments a few companies are conducting that outperform classical supercomputers, there has so far been nothing truly groundbreaking for banks in the area to look at practically. He believes quantum computing may have been a victim of its own projected success, which has led to current disappointment with timeframes.

“The progress of quantum computing has been simultaneously impressive, but also underwhelming from the perspective of listening to what the quantum computing industry said would happen in the past. But I would argue those were unrealistically optimistic expectations,” Sanches says. “I certainly would not say that [UBS’s decision] is a mistake or anything like that.”

Sanches says focusing on quantum computing does require a significant R&D commitment and that is being spearheaded by large banks with significant resources to spare, such as JP Morgan and Wells Fargo.

Scotiabank’s Jaffer says banks are also extremely careful not to over-invest in emerging technologies without an immediate financial benefit, noting that there is also doubt within the field about the advancements in quantum computing technology, given past over-optimistic predictions.

“The hardware guys are incredibly skeptical simply because that’s part of their business. They’ve overpromised and underdelivered for the last 10 years, so they’re trying not to do that again.”

If, or when, quantum computing has its ChatGPT moment, quantum evangelists are hoping it will increase the focus on the technology from the hypothetical advantages to real solutions.

Fujitsu’s Severini says the focus should not be on immediacy, but rather on careful preparation.

“If that is your aim, to achieve competitive advantage here and now, then the existing quantum computing machines cannot, at this moment, help you achieve that. What they can do is help you as an organization prepare for the point at which it will be possible to make use of it.”