Quantum Computing for Anti Money Laundering

In collaboration with Vanguard, this project investigates the use of quantum and hybrid quantum-classical techniques to enhance anti-money laundering (AML) analytics in large financial systems. Modern AML workflows analyze transaction histories, customer behavior, temporal patterns, and complex account-to-account relationships to detect potentially suspicious financial activity.

Conventional AML systems face several operational challenges as datasets grow in scale and complexity. Large transaction networks require intensive graph analytics and pattern mining, while evolving fraud behaviors introduce concept drift, requiring frequent retraining of detection models. These factors increase computational cost, pipeline latency, and the difficulty of maintaining interpretable detection frameworks.

The WISER research team is developing quantum feature maps and kernel-based learning models to provide a richer representations of high-dimensional financial data. A central focus of the effort is benchmarking the scalability and practical performance of hybrid quantum workflows for AML applications. Rather than replacing existing classical detection systems, the project evaluates how quantum-enhanced techniques could extend current pipelines in areas where classical methods begin to encounter computational or modeling limitations.

WISER Research Fellows: Shiplu Sarker, Vanshaj Kerni, Devanshu Shekhar, Nitesh Kansara, Varun Puram