It is time to build robust cross-industry anti-fraud and scam capabilities
Abstract
Recent years have seen a significant acceleration in the adoption of instant payment systems around the world, with about one-fifth of the global share of payments being processed instantly. These systems bring unprecedented speed and efficiency to the payments market, offering greater convenience for consumers. At the same time, however, they also enable fraudsters to operate more effectively. As a result, both the speed and volume of fraud have increased. In 2024 alone, global fraud losses were estimated to exceed US$1tn; almost 1 per cent of global GDP. The perpetrators are increasingly transnational criminal organisations using complex, multi-bank transaction schemes to conceal the destination of illicit funds. As a result, no single bank can gain full visibility of these networks through its own data alone. Standard rule-based and statistical approaches to fraud detection, relying on siloed bank-level data, are limited in effectiveness because they fail to capture network dimensions. This paper argues that the issue can only be addressed effectively through a holistic view of payment data at national and cross-border levels. This can be achieved by consolidating a shared data hub that enables: (1) real-time tracing and tracking of fund movements, allowing faster recovery for victims, quicker identification of mule accounts, and lower costs; and (2) more accurate fraud detection and risk scoring using graph-based data features. Moreover, when cross-industry and transnational utilities are built on such a hub, they enable mitigation and prevention strategies to operate synergistically, with each strengthening the other in a self-reinforcing cycle of resilience. This article is also included in the Business & Management Collection which can be accessed at http://hstalks/business.
The full article is available to subscribers to the journal.
Author's Biography
Federico Musciotto has more than 10 years experience in deep-tech analytics and academic research on network science and its applications. He leads and develops different solutions across the financial crime and national security sector, with a particular focus on network anomaly detection and graph AI models for fraud detection on faster payment systems, contributing to dozens of network graph analytics projects across both developed and emerging market jurisdictions.
Matteo Neri is a lead data scientist at FNA. He has more than eight years of experience, focused on deep-tech analytics and data science. A physicist and network scientist by training, he leads and develops solutions across the national security and private sectors, with a particular focus on network anomaly detection, entity resolution and Big Data solutions. He co-leads the development of AI-driven tools for detecting consumer fraud, as part of FNA’s National Fraud Portal initiative.
Kimmo Soramäki is a recognised thought leader in financial network analytics and simulation, and the founder and Chief Executive Officer of FNA. Kimmo has advised and collaborated with central banks, regulators and international organisations for over 25 years. His academic work has influenced global discussions on financial infrastructure resilience, data-driven supervision, and the future of systemic risk oversight. Under his leadership, FNA has grown into a trusted partner for over 50 central banks, financial market infrastructures, large global banks and international organisations worldwide. Kimmo contributes regularly to World Bank, International Monetary Fund, Bank for International Settlements and G20 events, as well as to industry working groups, and continues to drive FNA’s mission to make the financial system safer and more efficient through innovative, explainable and collaborative technology. Kimmo holds a doctor of science in operations research and a master of science in accounting (finance), both from Aalto University in Helsinki.
