Global: Swift, Global Banks Trial AI and Data-Sharing Tools to Combat Financial Fraud

Swift, in collaboration with leading international banks and technology partners, has successfully completed a series of experiments using artificial intelligence (AI) and privacy-enhancing technologies (PETs) to strengthen the global fight against financial fraud.

The trials demonstrated how PETs can allow financial institutions to securely share fraud-related intelligence across borders without exposing sensitive customer information. In one use case, participants were able to verify suspicious account activity in real time—significantly reducing the time it takes to identify cross-border fraud networks and prevent fraudulent transactions from being executed.

In another test, participants combined PETs with federated learning, an AI technique that enables models to train locally on each institution’s data. This approach allowed banks to collaborate without transferring underlying customer data, protecting privacy while enhancing fraud detection.

Trained on a dataset of 10 million synthetic transactions, the federated learning model was shown to be twice as effectiveat detecting known fraud cases compared to models trained on a single institution’s dataset.

Financial institutions that participated in the trials included ANZ, BNY Mellon, and Intesa Sanpaolo, alongside technology partners such as Google Cloud.

Rachel Levi, Head of AI at Swift, highlighted the potential of the initiative:

“The industry loses billions to fraud each year, but by enabling the secure sharing of intelligence across borders, we’re paving the way for fraud to be stopped in minutes rather than hours or days.”

Following the success of the first phase, Swift plans to expand the scope of participation and launch a second round of trials using real transaction data to demonstrate the technologies’ impact in real-world conditions.

The initiative is part of Swift’s broader efforts to explore AI applications in cross-border payments. The cooperative is currently working on over 50 AI-driven use cases, ranging from proofs of concept and pilot projects to live services.

Earlier this year, Swift also launched its AI-enhanced Payments Controls Service, designed to help small and medium-sized financial institutions more effectively identify and block suspicious transactions in real time.

By advancing AI-driven collaboration and secure data sharing, Swift and its partners are laying the groundwork for a more resilient, transparent, and fraud-resistant global payments ecosystem.