By Mackenzie Tatananni

As the nation's largest bank, JPMorgan Chase seems like an unlikely pioneer in quantum computing.

But it is forging ahead in the sector after announcing that it helped achieve a milestone by demonstrating a real-world application of a quantum computer.

In a peer-reviewed paper published in the journal Nature last month, researchers claimed they had achieved certified randomness. This came after researchers at JPMorgan Chase wrote an algorithm to generate random numbers, which ran on a Quantinuum computer.

Leading the charge for JPMorgan is Marco Pistoia, the bank's head of global technology applied research. His hiring in January 2020 led to the formation of the first applied research team, but Pistoia recalls how JPMorgan showed an interest in quantum long before then.

While working at IBM Research in 2018, Pistoia heard JPMorgan was hiring engineers and mathematicians to test IBM's quantum computer prototypes despite the technology being in its infancy.

"Back then, I was like, wow, this is so exciting, a company like JPMorgan is already interested in quantum computing," Pistoia told Barron's. Just a few years later, he found himself "on the other side," spearheading efforts to conduct research across several fields of technology.

Pistoia concedes quantum computing might seem like a departure from what JPMorgan is best known for. However, a lot of banks are in the arena today, including Citigroup, HSBC, and Wells Fargo. HSBC, for one, has embraced so-called disruptive technologies, even claiming in March 2024 that quantum's "ChatGPT moment" was on the horizon.

Banks must use the tools at their disposal today, but classical computers aren't a perfect fit. Quantum is positioned as a better alternative in tasks such as risk modeling and analysis, as well as "a lot of things in the domain of machine learning," Pistoia explained.

In fact, the JPMorgan team has already developed algorithms that dramatically reduce the time needed to train a large language model. But there's a catch.

"Of course, the algorithms aren't yet usable in production because the quantum computers aren't yet powerful enough to support them," Pistoia said.

The bank doesn't build computers themselves, but provides financial support to companies that do. In the latest experiment, researchers remotely accessed a Quantinuum System Model H2 and compared its performance to supercomputers housed at federally-funded laboratories across the country. By completing a number-generation task beyond the capabilities of classical machines, the researchers were able to declare quantum supremacy, weeks after D-Wave Quantum made similar claims with regard to a materials simulation problem.

It might seem like nothing more than a way to add another feather to JPMorgan's cap, but random numbers are essential for the generation of cryptographic keys and other security-related parameters that keep digital information safe from prying eyes.

The organization knows firsthand the devastating effect of a data breach. In 2014, a massive cyberattack compromised the accounts of 83 million clients after JPMorgan spent months trying to contain it.

The threat landscape is only expected to get more treacherous with the advent of quantum, as cybersecurity experts predict quantum computers will be able to crack algorithms that have protected digital information for decades.

JPMorgan is working to solve this problem, but success won't happen overnight. Scott Aaronson, the computer scientist who first proposed the certified randomness protocol in 2018, noted that the results, while promising, weren't yet good enough for "high-stakes cryptographic applications."

Quantum development requires continuously pushing the boundaries of science and engineering, making it difficult to pin down a timeline, but Pistoia thinks commercially viable quantum computers could arrive by the end of the decade.

"We're upfront about the limitations of our work as well as the good qualities," he said. "We've taken a step towards a commercial use of quantum computing, but it's not commercial yet. We're getting there."

Write to Mackenzie Tatananni at mackenzie.tatananni@barrons.com

This content was created by Barron's, which is operated by Dow Jones & Co. Barron's is published independently from Dow Jones Newswires and The Wall Street Journal.

(END) Dow Jones Newswires

April 08, 2025 12:19 ET (16:19 GMT)

Copyright (c) 2025 Dow Jones & Company, Inc.