How Quantum Computers are Transforming Banking through Financial Engineering?

In the era of rapid technological development, quantum computers are gaining increasing importance in many fields, including the banking sector.

The introduction of quantum computers in banking can revolutionize the way financial institutions process data, analyze risk, and make strategic decisions. Major banks like Goldman Sachs and J.P. Morgan are conducting advanced research related to this technology. However, work is also being done in this field in Poland.

We had the opportunity to talk to experts. Rafał Pracht and Marek Stefaniak from BNP Paribas Bank told us about the progress of work on quantum computers and the potential benefits and challenges associated with their implementation.

This will help understand how this innovative technology can impact the future of finance and how banks are preparing for this groundbreaking step.

Web Economy: How Will Quantum Engineering Change Banking?

Rafał Pracht: Currently, the main application of quantum technology is in the field of security. The Shor’s algorithm theoretically allows breaking the RSA algorithm.

Web Economy (editor’s note): The Shor’s algorithm is a quantum algorithm used for factoring numbers into prime factors. In theory, it can break the RSA algorithm, which relies on the difficulty of factoring large prime numbers for its security. If a quantum computer could execute the Shor’s algorithm, RSA-based systems would become vulnerable to attacks.

Rafał Pracht: However, already this year, the NSA, the U.S. National Security Agency, demonstrated algorithms that are resistant to being broken by the Shor’s algorithm. In our bank, we focus on the field of financial engineering. We want to use quantum computers to solve problems that are very complex or even impossible to solve on classical machines, such as pricing options or other more complex financial instruments.

Web Economy: Are you working on projects in the bank that can respond to the new possibilities of quantum engineering?

Rafał Pracht: Yes, we have managed to publish two articles. In one of them, we show how to apply quantum computers to price Asian options, and in the other, we present a way to load a distribution onto a quantum computer.

Web Economy: Loading a distribution onto a quantum computer is the process of initializing the quantum state of the computer to correspond to a given probability distribution. In practice, this means preparing qubits in such a way that their probability amplitudes (in the quantum state space) correspond to the probability values in the given distribution.

Loading a distribution onto a quantum computer is essential in many quantum algorithms, such as Grover’s algorithm or quantum structured search algorithm. In these cases, appropriately preparing the quantum state at the beginning of the computation allows us to take advantage of quantum properties, such as superposition and entanglement, to speed up information processing and achieve results that cannot be obtained with classical computers.

Rafał Pracht: It’s not an easy task, and we need a large number of operations on a quantum computer, which can make them inefficient. We have developed an algorithm that is very efficient, and we expect that once such machines are ready, we will be able to gain a competitive advantage by utilizing our solutions in the cloud.

Marek Stefaniak: Rafał mentioned one engineering aspect, but what we focus on is the educational dimension. We talk to the business, our product owners, to teach them about this technology and help them understand the possibilities that quantum computers create for their specific products. This is also an important aspect of our activities. We pursue two paths: one is purely technological, where we develop concrete solutions, and the other is the awareness and educational aspect, in which we also specialize.

Web Economy: What are people’s reactions to this technology?

Marek Stefaniak: I think there are two perspectives on this issue. Many people are genuinely interested because it is a novelty and potentially brings many additional competitive advantages. Fortunately, we have identified several individuals in our bank who are genuinely interested in applying this technology in their business areas. There is also another personality profile, people who are not convinced by novelties and prefer to focus on technologies and solutions they are familiar with. We have to navigate in an environment of both these groups of people.

Web Economy: How many more years does this technology need to reach maturity?

Rafał Pracht: If we look at the map of technology providers, for example, there are claims that within three to four years, 256-qubit quantum computers will be available. To understand how big that is, it is generally accepted that 65-70 qubits are a size that cannot be simulated on any computer in the world. So, 256 qubits is well above that limit. However, these are low-quality qubits, so it is uncertain how much we will be able to utilize them.

On the other hand, error correction algorithms have already been developed, and IBM is another provider. They already have a 127-qubit computer, but of much lower quality. However, they expect to be able to correct errors. They even have plans for machines with 1000 qubits in two to three years. If we add the overhead associated with error correction, in our estimation, in 4-5 years, machines may appear that can be used, perhaps not to demonstrate the impossibility of performing computations on a given computer, but at a level comparable to servers with graphics cards (GPUs).

Marek Stefaniak: This year, a report was released in which global companies that use quantum computers were surveyed. 43% of respondents expect that within the next year to three years, the first commercial applications of this technology will appear. So it seems that a period of 25 years is overstated, and it will actually be a period from one to five years.

Rafał Pracht: Especially since we already have solutions where quantum computers are better than classical computers. In neural networks, with GANs (editor’s note: generative adversarial networks), where we generate some images, Zapata Computing showed an algorithm last year that can generate better images than what ordinary neural networks can do. Actually, this quality comes from randomness, and even errors don’t hinder quantum computers; they actually help. These small disturbances cause the randomness and quality of these images to be at a high level.

Web Economy: Will quantum computers have an impact on personalized offers?

Marek Stefaniak: Yes, I believe so. It is the direction of product development, where greater computing power combined with artificial intelligence will enable even greater customization and adaptation of products to customer needs.

Web Economy: What technical threats can we expect?

Rafał Pracht:  Currently, all quantum computer providers operate in the cloud. Security will be at the same level. If we consider the cloud to be secure, then quantum computers will be equally secure. The difference will not be noticeable.