Swift and Google Cloud combat fraud with federated learning

Artificial intelligence innovation. AI adoption and working. AI with a digital brain in a lightbulb with human icons in wooden blocks for AI vs human competition. manage personnel in the organisation.
Editorial credit: chayanuphol / Shutterstock.com

Swift has partnered with Google Cloud to create anti-fraud technologies that leverage advanced AI and federated learning.

Under the agreement, Swift is teaming up with Google Cloud, along with partners Rhino Health and Capgemini, to create a solution for financial institutions to fight fraud. The approach combines federated learning with privacy-enhancing technologies to enable data sharing while protecting sensitive information.

Rachel Levi, Head of AI at Swift, commented: “Swift is in a unique position in the financial industry – a trusted and cooperative network that is integral to the functioning of the global economy. As such, we are ideally placed to lead collaborative, industry-wide efforts to fight fraud. 

“This exploration will help the community validate whether federated learning technology can help financial institutions stay one step ahead of bad actors through sharing of fraud labels, and in turn enabling them to provide an enhanced cross-border payments experience to their customers.”

This approach aims to address major challenges faced by financial institutions, such as limited data availability due to concerns and regulations around data sharing. 

Federated learning looks to provide a solution by allowing collaborative AI model training while safeguarding privacy. Rather than collecting sensitive data from financial institutions, the model is trained on decentralised data within each institution.

Andrea Gallego, Managing Director of Global GTM Incubation at Google Cloud, said: “At Google Cloud, we are committed to empowering financial institutions with cutting-edge technology to combat the evolving threat of fraud. 

“Our collaboration with Swift exemplifies the transformative potential of federated learning and confidential computing. By enabling secure collaboration and knowledge sharing without compromising data privacy, we are fostering a safer and more resilient financial ecosystem for everyone.”

Additionally, in the first half of 2025, Swift will launch a sandbox using synthetic data to prototype learning from past fraud. 

The initiative will involve 12 global financial institutions, with Google Cloud as a key partner. 

This effort builds on Swift’s existing Payment Controls Service (PCS) and follows a successful pilot with firms from Europe, North America, Asia and the Middle East.