For most people in institutional finance, quantum computing is an entirely abstract concept.
In reality, the idea is simple enough: quantum computers use the principles of quantum mechanics to process information in ways that today’s machines cannot.
Classic computers store data as bits – that is zeros and ones. Quantum computers use qubits, which can be both 0 and 1 at the same time.
This property, known as superposition, allows quantum machines to explore many possible outcomes simultaneously rather than sequentially.
Another difference is something called entanglement. Entanglement lets qubits influence one another instantly, enabling patterns of correlation that classical systems cannot replicate.
Should you care?
Truthfully, for most everyday tasks, this makes no difference whatsoever. But for certain categories of problems, involving vast combinations, complex probability distributions or extremely high‑dimensional search spaces, quantum computers could, in time, offer dramatic speed‑ups.
That is why the technology has captured the attention of the world’s largest banks.
Institutional capital markets handle intense problems. Whether it is pricing exotic derivatives, simulating risk across thousands of scenarios, optimising portfolios under tight constraints, or securing global transaction flows, there are certainly reasons to care.
Banks are investing in exploring quantum because even marginal improvements in optimisation, simulation or security can translate into a material financial advantage. In the case of cryptography (which some have only just got their heads around) it could also prevent catastrophic downside.
What began as a research curiosity is now a strategic capability.
Several globally significant institutions, such as JP Morgan, Goldman Sachs, HSBC, BNP Paribas and Barclays are already conducting work that signals a new phase.
Externally, they are beginning to publish research to market that is concrete, empirical and crucially, candid about what quantum can and cannot do.
A review of public work from these five banks reveals a set of emerging themes. They are not the breathless claims of a technology bubble, but the measured assessments of institutions accustomed to pricing risk.
Quantum for optimisation
Most financial problems, portfolio construction, asset allocation, collateral optimisation, are optimisation problems. They involve searching through vast combinations of possibilities to find the best outcome under constraints. Classical computers can do this, but the cost grows exponentially with complexity.
Quantum computers, in theory, can explore these combinations more efficiently.
JP Morgan has arguably been the most active bank publishing technical work. In 2024, its Global Technology Applied Research group released a study on quantum‑enhanced portfolio optimisation using Quantinuum’s trapped‑ion hardware.
It concluded: “Portfolio optimisation is an important use case in finance that lends itself to be tackled by quantum computing.”
The team introduced “Hybrid HHL++”, a compressed version of a well‑known quantum algorithm for solving linear systems, describing the experiment as: “the largest experimental demonstrations of HHL for an application to date.”
While the message was understated, it was clear that quantum hardware is still small and noisy, but careful engineering can already push it into financially meaningful territory.
The Monte simulation
In technology, a Monte Carlo simulation is a way of estimating the value of something by running a very large number of random scenarios and seeing how the outcomes average out.
In finance, it’s used when a payoff or risk measure is too complex to calculate directly, so you simulate thousands or millions of possible market paths, apply the payoff rules to each one, and use the distribution of results to estimate the price or the risk.
Many of the hardest problems in capital markets boil down to running huge numbers of simulations.
Whether you are pricing an exotic derivative, calculating XVA, or running VaR and stress tests, you are essentially firing off millions of Monte Carlo paths and averaging the results. It works, but it is slow and expensive. Even with modern hardware, these simulations can dominate overnight batch runs and intraday risk cycles.
Quantum computing matters here because, in theory, certain quantum algorithms can reduce the number of simulations needed. One of the most promising is called amplitude estimation, which could deliver the same accuracy with far fewer simulation steps.
Goldman Sachs decided to ask a very practical question: how big would a quantum computer need to be before it could price derivatives faster than the best classical systems?
In a study entitled A Threshold for Quantum Advantage in Derivative Pricing, the conclusion is deliberately blunt. To beat classical machines on complex structures such as autocallables, a quantum computer would need abilities that are way in excess of today’s technology – around 8,000 logical qubits or a circuit depth of around 54 million operations.
The researchers concluded: “While the resource requirements… are out of reach of current systems, we hope they will provide a roadmap.”
Goldman’s research effectively means that quantum could one day transform derivative pricing, but the machines capable of doing so do not exist yet. When they do, the first wins will likely be in very specific, high‑value pricing problems.
But do not be misled. Banks that want to benefit from quantum advantage later must build internal expertise now.
Quantum computing is not about to rewrite the rules of finance. But nor is it a curiosity.
The world’s largest banks are treating it the way they treat any emerging asset class: with caution, capital and a clear understanding that timing matters.
If quantum advantage arrives, even in narrow domains, the institutions that built capability early will be best placed to capture it. And if it arrives later than expected, the investment will still have bought something valuable: resilience, security and optionality.
