The Alan Turing Institute, Plenitude Consulting, Napier AI and the Financial Conduct Authority (FCA) today announce a new project aimed at improving money laundering detection through the creation of a new fully synthetic dataset, by using real anonymised financial transactions from high street banks augmented with a wide range of money laundering typologies.
Findings by the FCA show that access to sufficiently realistic data is a key barrier for developing and demonstrating the effectiveness of alternative and innovative approaches to money laundering detection beyond typical “rules based” approaches.
The project will address the current barriers to innovation by creating a new synthetic dataset that will be available in the coming months via the FCA’s digital sandbox. In an isolated testing environment, firms developing new and emerging techniques to detect money laundering will be invited to apply to use the synthetic data sets to train, test and demonstrate new approaches and algorithms for money laundering detection.
The participants will have the opportunity to explore and use the dataset to demonstrate the effectiveness of their money laundering detection solutions, whilst also measuring the dataset’s effectiveness. Synthetic data that mimics the characteristics of real-world data can help businesses test their models and systems in a variety of scenarios to make sure that they are robust and reliable.
The project is part of a greater effort by the FCA to leverage analytics and data science to improve efforts to tackle financial crime, reducing harm and increasing trust in financial services. The project is aligned with the goals of the UK Economic Crime and FCA’s Business Plan, to create a more competitive and dynamic market for money laundering detection solutions and increase the ability and the efficacy of retail banks to detect and prevent money laundering.
The Turing and FCA have previously collaborated on reports exploring the impact of emerging technologies on the financial services industry and ‘Using Synthetic Data in Financial Services’ as part of the FCA’s Synthetic Data Expert Group.
Professor Lukasz Szpruch, Director for Finance and Economics at The Alan Turing Institute said: “This collaboration between the Turing, Plenitude, Napier AI and the FCA will enable UK retail banks and solution providers to develop and evaluate the effectiveness of new approaches, solutions and money laundering detection algorithms, and asses the use of synthetic data in this context. Ultimately, the project will equip banks and solution providers with invaluable tools to combat money laundering and protect their customers.”
Top image credit: weerapat1003 via Adobe Stock