Belgium-headquartered Society for Worldwide Interbank Financial Telecommunications (SWIFT), the global specialist in financial transactions, is launching an artificial intelligence (AI)-driven pilot in partnership with Axis Bank to help combat payments fraud occurring in cross-border payments. Along with global banks, including BNY Mellon, Deutsche Bank, DNB, HSBC, Intesa Sanpaolo and Standard Bank, the duo will test the use of secure data collaboration and federated learning technologies.
The pilot is expected to leverage a secure infrastructure that will enable financial institutions to exchange relevant information with strong privacy-preserving controls, according to the two partners. SWIFT’s AI anomaly detection model will then be able to gather insights and identify potential fraud patterns from a much richer dataset, they said.
“For the purpose of fraud control, this pilot looks at how we can incorporate newer technologies like artificial intelligence into the algorithms that we have,” said Neeraj Gambhir, group executive, treasury, markets and wholesale banking products, Axis Bank. “There is a static rule-based algorithm that SWIFT implements. This latest initiative will add another layer to that, using AI and machine learning features to make it more dynamic. We are the only bank chosen from India to see how we can incorporate AI, machine learning models into fraud detection engines that are currently running on the SWIFT platform.”The pilot will use AI algorithms to analyse historical transaction data and detect anomalies indicative of fraud. The real-time and data-driven approach aims to refine the model for broader application across SWIFT’s network to enhance the accuracy and efficiency of fraud detection.
The new AI-enabled system will facilitate enhanced monitoring and reporting, allowing financial institutions to track transactions in real-time and generate detailed reports. These capabilities not only aid in identifying and understanding fraudulent activities but also contribute to increased consumer trust by ensuring the safety and integrity of their transactions.
“To tackle the pressing issue of fraud, we are developing innovative solutions, including AI technologies, that enhance fraud detection,” said Kiran Shetty, CEO, SWIFT, India & South Asia. “AI’s ability to look at richer and larger datasets will bring the financial community one-step ahead in a world of rising cyber risks.”
According to the Reserve Bank of India, fraud cases in the banking sector increased nearly 300% over the past two years, reaching 36,075 cases in 2023-24. In 2023, according to the Nasdaq-Verafin Global Financial Crime Report, global fraud incurred a staggering cost of $485 billion.
SWIFT is also working towards contributing to the cross-border interoperability of potential central bank digital currencies (CBDCs).
“The organisation’s experiments have demonstrated that CBDCs can be interlinked through a single gateway, ensuring security, compliance, authentication and tracking within the SWIFT network,” said Shetty.
Currently, the organisation allows members to transact about 150 currencies, 40,000 corridors and more than 200 countries and territories.
