The International Monetary Fund is advocating for banks to reassess their traditional reservations about sharing data among financial institutions in the realm of financial crime prevention. In a recently released Technical Note, it emphasizes that disjointed information systems are detrimental to combating AI-enabled fraud. The document encourages greater collaboration by urging institutions to share private information.

This Technical Note was issued during the 2026 Spring Meetings of the IMF and focuses on proactive strategies for financial institutions to tackle digital fraud more effectively. It highlights that efforts to counter such fraud have been compromised due to banks’ hesitance to exchange threat data, both domestically and internationally.

Artificial Intelligence (AI) has empowered criminals by enabling them to gather large volumes of data, which they use in sophisticated attacks. In response, the IMF is pushing for a more cooperative approach, particularly advocating the sharing of transactional and threat data across different institutions.

Crucially, the report warns that technological advancements alone are insufficient; they should be paired with comprehensive data-sharing practices, especially concerning transaction records, to enhance the collective ability of financial institutions (FIs) in detecting, preventing, and mitigating illicit finance activities.

More Data, More Value

AI and machine learning are particularly adept at identifying suspicious transactions; however, their effectiveness hinges on access to extensive, varied datasets. The Technical Note identifies siloed data architectures as the main barrier in fighting fraud effectively.

In contrast, AI tools perform better in integrated systems that leverage shared data sets, facilitated by application programming interfaces (APIs) and common standards such as ISO 20022. The IMF underscores the importance of APIs, standardized data formats, and interoperability frameworks to promote meaningful data exchange among financial institutions.

BREAKING DOWN THE SILOS

While there are valid concerns about data sharing due to competitive pressures and regulatory constraints, the growing complexity and global nature of fraud networks underscore the need for increased transparency and collaboration. This approach could significantly strengthen the financial system’s capability to detect and prevent illicit activities.

“Data sharing and collaboration remain a significant challenge within financial institutions, not only between banks but also across different lines of business within the same organization,” noted Suzanne Sando, Lead Analyst of Fraud Management at Javelin Strategy & Research. “A financial institution might identify signals to halt fraud on a customer’s credit card but may lack the critical risk signals and emerging threat trends needed to prevent fraud on that same customer’s debit account. These data silos prevent banks from accessing essential information necessary to keep up with fraudsters, especially as AI continues to evolve and be used by criminals.”