finsmart-news_admin Contents
How Swift’s AI Experiments Work
Swift, the cooperative at the center of global financial messaging, has revealed the results of groundbreaking experiments designed to fight fraud in international payments. The initiative involved 13 financial institutions working together to test the use of artificial intelligence and privacy-enhancing technologies (PETs) in fraud detection. By allowing secure sharing of fraud-related insights across borders, these experiments aimed to reduce losses and accelerate the speed at which fraudulent activity can be intercepted.
One of the key use cases involved verifying intelligence on suspicious accounts in real time. PETs allowed participants to share insights securely, enabling them to identify potential financial crime networks before fraudulent transactions were executed. This represents a shift away from siloed intelligence and toward cooperative fraud defense across borders.
In another test, federated learning was applied to synthetic datasets representing 10 million artificial transactions. The model trained locally at each institution without exposing customer data. Compared to single-institution models, this collaborative approach was twice as effective in detecting known fraud patterns, suggesting powerful applications for cross-border fraud defense at scale.
Takeaway
Federated learning and PETs doubled fraud detection effectiveness, proving the value of secure, cross-border AI collaboration in financial services.
Why Cross-Border Collaboration Is Crucial
Fraud in cross-border payments remains a pressing challenge, with financial crime costing the industry an estimated $485 billion in 2023. Swift’s experiments highlight that fragmented responses by individual banks cannot match the scale of fraud networks operating across jurisdictions. A unified approach, leveraging AI and privacy tools, could significantly reduce detection times and industry-wide losses.
Rachel Levi, Head of AI at Swift, emphasized the power of collective action: “These experiments demonstrate the convening power of Swift as a trusted cooperative at the heart of global finance. A united, industry-wide fraud defence will always be stronger than one put up by a single institution acting alone.” By enabling institutions to exchange intelligence in minutes, not days, Swift is paving the way for a faster, more secure global payment system.
Swift has been actively exploring AI applications, with more than 50 projects in pilot or production, including the AI-enhanced Payments Controls Service launched earlier this year. That service already helps small and medium-sized banks flag suspicious transactions in real time, underscoring Swift’s ongoing push to embed AI in its offerings.
Takeaway
Fraud is a $485 billion problem—cross-border intelligence sharing can cut detection times dramatically and strengthen global payment security.
What Participants And Partners Are Saying
Several major institutions and technology providers took part in the experiments, including ANZ, BNY, Intesa Sanpaolo, and Google Cloud. Their involvement underscores the broad industry appetite for collaborative solutions that protect consumers and institutions alike. The feedback from participants suggests optimism about AI-driven fraud defense becoming a core component of future payment infrastructures.
David Buckthought of ANZ stated: “The rise in fraud and scams is a global issue impacting all financial institutions. ANZ is excited to be involved in an industry-wide response, proving the use of federated learning to enhance detection capabilities. This will provide banks with a stronger defence against fraudulent activity.” His remarks highlight the importance of collaboration across competitive institutions to create systemic resilience.
Echoing the sentiment, Isabel Schmidt of BNY said: “Security is paramount in cross-border payments. Using the latest technologies, this group has achieved results that show how these tools can be used to uplift the entire ecosystem and demonstrate the value of the Swift co-operative in bringing competitive organizations together behind a greater good.” Meanwhile, Enrico Canna of Intesa Sanpaolo added: “Fraud in cross-border payments increases friction in the ecosystem and causes significant costs at an industry level. Intesa Sanpaolo is collaborating in these preliminary experiments led by Swift to demonstrate the positive impact of a synergistic approach supported by the latest technologies for making the ecosystem more secure and reliable.”
Takeaway
Global banks back Swift’s AI-driven approach, citing stronger security and a cooperative model that benefits the entire payments ecosystem.
What Comes Next For Swift’s Fraud Program
Following the success of the initial trials, Swift plans to expand participation and move toward real-world testing with live transaction data. The second phase will examine how PETs and federated learning models perform in operational environments, where speed and accuracy directly impact financial outcomes. If successful, the technology could enable banks worldwide to block fraudulent payments in real time.
Industry observers see this as a major step forward in the modernization of fraud defenses. By leveraging AI responsibly, with guardrails for privacy and compliance, financial institutions can better protect customers without compromising sensitive data. Such a system could also ease regulatory concerns by offering more consistent fraud detection standards across jurisdictions.
The long-term vision is a cooperative fraud shield built on shared intelligence, real-time data processing, and AI models capable of evolving with new threats. For global finance, this could mark the beginning of a new era where fraud detection is not the responsibility of one institution but a shared mission across the entire ecosystem.
Takeaway
Swift’s next step is real-world testing, aiming to turn its AI experiments into a live cross-border fraud defense system for the global banking industry.
Previous articleDotfile Debuts Autonomy, A Self-Decisioning KYB AI
