Exploring quantum computing at Lloyds Banking Group

Quantum computing is rapidly emerging as one of the most transformative technologies of the next decade, offering the potential to tackle computational challenges beyond the reach of classical systems. 

At Lloyds Banking Group, we are committed to understanding how such breakthrough technologies could shape the future of financial services and enhance the outcomes we deliver for our customers. One of our most significant steps on this journey has been a nine-month quantum computing experiment carried out in collaboration with IBM, exploring how quantum algorithms might support economic crime prevention. 

This article summarises the experiment, the collaborative approach behind it, and what the findings mean for our long-term innovation agenda.

Why quantum, and why now?

Financial services organisations face increasingly complex computational challenges, from fraud detection to optimisation and simulation tasks. While artificial intelligence and classical machine learning continue to be essential tools, quantum computing may offer advantages in situations where traditional architectures reach their limits. 

Early research by our Chief Technology Office team highlighted several areas where quantum computing could outperform classical methods, including graph based anomaly detection; an important capability in understanding fraudulent or criminal behaviours. This formed the foundation for choosing economic crime prevention as the first domain for experimentation. 

Lloyds Banking Group and IBM: a collaboration built on exploration

The collaboration with IBM brought together subject matter experts from across both organisations. The project unfolded across a discovery phase – mapping quantum opportunity areas across the Group – and an experimentation phase focused on one specific challenge in economic crime prevention: detecting mule activity using graph analytics. 

This involved combining anonymised real transactional data with quantum algorithms executed on IBM’s cloud based quantum computers. The goal was not to deliver production ready solutions, but to understand which quantum techniques show genuine long-term promise.

We explored multiple quantum algorithmic approaches, including techniques that had not been widely tested in this context. While results should not be overstated, several approaches showed promising early behaviour, and the experiment is considered among the largest of its kind conducted on real quantum hardware.

The computational challenge, and why quantum matters

Economic crime prevention, particularly the detection of mule accounts, requires analysing highly complex networks of financial transactions. These can be represented as graphs of customers, accounts, and payments, where suspicious activity often hides in subtle network structures.

Traditional computers struggle with certain classes of graph problems because the number of possible solutions grows exponentially with problem size, making them among the most challenging problems to solve using classical computation. Quantum computing holds the potential to mitigate these limitations by exploring vast solution spaces more efficiently than classical hardware. 

Our experiment did not aim to explore how to replace machine learning models currently used in fraud and crime prevention. Instead, it explored whether quantum enhanced techniques could one day generate more sophisticated graph-based features to support future models; features that might be too complex or expensive to compute classically.