Collaboration in anti-money laundering (AML) is widely acknowledged as essential, yet progress remains limited. While credit reference agencies already share data across industries and fraud reporting hubs exist in some regions, cross-institutional AML cooperation often stalls before reaching operational reality.
According to Consilient, regulators, financial intelligence units (FIUs) and banks agree that financial crime is a shared problem requiring a shared solution. Initiatives such as the UK’s National Crime Agency (NCA) Data Fusion programme and the Netherlands’ TMNL have shown the potential of joint efforts. However, most projects never move beyond policy statements or pilots, hampered by legal, operational, and technological hurdles.
One of the biggest blockers is regulatory misalignment. Although the Financial Action Task Force (FATF) provides a global AML/CFT framework, interpretation and enforcement vary significantly between jurisdictions, complicating data sharing. Without clear regulatory backing, collaboration often feels riskier than acting alone.
Data privacy concerns also loom large. Laws such as GDPR and CCPA impose strict rules on information use and retention. Even anonymised data can present risks, and many institutions prefer to retain control rather than expose themselves to reputational or regulatory consequences.
Internal fragmentation adds further complexity. Compliance, legal, risk and technology teams often work in silos, with conflicting priorities and systems that don’t communicate effectively. For multinational banks, data residency laws and local governance frameworks can make even internal cross-border sharing a challenge.
Technical infrastructure is another stumbling block. Many organisations lack the tools to share or integrate data securely, with outdated systems and incompatible formats creating operational friction. On top of this, competitive pressures mean institutions can be reluctant to share insights they see as a potential advantage.
Some attempts have succeeded. In Singapore, the Monetary Authority of Singapore’s AML/CFT Industry Partnership (ACIP) brought banks, regulators and law enforcement together to address specific typologies, supported by clear legal guidance. The UK’s NCA Data Fusion programme has also worked by focusing on high-impact use cases and avoiding wholesale data pooling.
Others, such as SWIFT’s Transaction Monitoring Utility, the Nordic KYC Utility, and blockchain-based KYC pilots, failed due to governance disputes, legal uncertainty and technical challenges. TMNL, despite early promise, is winding down operations to redesign its model for compliance with the incoming EU Anti-Money Laundering Regulation (AMLR).
A promising alternative is federated learning (FL), which enables institutions to train models locally on their own data, sharing only encrypted updates rather than raw records. This retains full data control, improves detection by incorporating behavioural patterns from multiple sources, and meets audit and regulatory requirements. Early adopters have reported a fourfold uplift in detection and a 75% improvement in analyst efficiency.
The lesson is clear: AML collaboration succeeds when it is private by design, outcome-driven, and operationally feasible. Federated learning meets these criteria, offering a model that aligns with today’s legal, technical and organisational realities — and finally turning shared intent into shared action.
For more, find on RegTech Analyst.
Copyright © 2025 FinTech Global
