Is the financial sector posed to be an early adopter of quantum computing technologies? An increasing number of industry observers believe this to be the case.
Historically, this sector is no stranger to applying physics to solve complex problems. Take, for example, the Black-Scholes-Merton model, which “uses the concept of Brownian motion to price financial instruments—like European call options—over time.”
We spoke with Aram Harrow, MIT Professor of Physics, to get his insights on the potential of quantum computing in finance.
Professor Harrow explains that many financial organizations are in the early exploratory phases of adopting quantum computing.
“Because it takes time to determine how quantum computers will be useful and how they can connect to systems, companies want to get started now before there are large-scale quantum computers. So currently, they’re in an exploratory mode where they’re testing quantum computing on small problems to gain clarity on its future uses,” he says.
There are many potential use cases for quantum computing in finance, such as “targeting and prediction, trading optimization, and risk profiling.” When asked where he believes quantum computing could make the biggest impact in this sector, Professor Harrow focused on three key areas: machine learning, secure communication, and risk management.
Professor Harrow sees the broad potential for quantum computing to improve machine learning across many sectors, including financial services. “Quantum computing could allow for better training of complicated machine learning models that can’t be trained effectively on classical computers,” he explains.
Regarding secure communication, Professor Harrow says, “There's a possibility of using quantum key distribution (QKD), and some organizations have been experimenting with that. QKD can play a significant role in a communication network when financial organizations really want to be sure about security.”
While Professor Harrow shares that there hasn’t been much practical work done yet in the area of risk management, he notes, “I think what's possible is that by looking at a more complicated model, financial services organizations might be able to consider a model that looks at a broader range of possibilities, which could include the type of unlikely events that could be risky scenarios they want to avoid.”
McKinsey reports that “quick wins for quantum computing [in finance] are most likely in areas where artificial intelligence techniques such as machine learning have already improved traditional classification and forecasting.”
When asked about the role AI plays in the potential adoption of quantum computing in finance, Professor Harrow noted that there are two different lenses from which to consider this question: quantum computing for AI and AI for quantum computing.
“If you’re going to build a quantum computer, then as it gets big, you might have a complicated system with patterns of errors that can be difficult to describe. In these cases, people have found that AI models are helpful for understanding and determining how to correct errors,” Professor Harrow explains. He adds, “AI can also be used to improve how quantum computers are built and operated.”
“The hope here is to build quantum computers that improve existing AI models,” says Professor Harrow. “What that will probably look like is some kind of hybrid computing where quantum and classical computers are used together to solve a problem. Researchers are trying to determine how to divide a problem to make the best use of resources on both types of platforms.”
Professor Harrow shares a few strategies financial leaders can use to prepare for quantum computing.
Learn more about the career impact of learning quantum computing online and enroll in MIT xPRO’s Quantum Computing fundamentals course to join other finance leaders paving the way for the future.