Fraudsters are exploiting the vulnerabilities of increasingly interconnected economies, leaving banks facing damaged customer relationships, intense regulatory scrutiny, and heightened reputational risk.
Near-instant payment settlement leaves little time for traditional fraud detection procedures, adding to a costly and growing challenge. While human analysts are grappling with a complex payments landscape, suspicious activities are staying under the radar of legacy rules-based systems. Artificial intelligence is becoming paramount to improving detection of fraudulent transactions, yet low-quality data is hampering model performance.
In a bid to improve the accuracy of their AI models, innovative banks are turning to secure intelligence sharing on fraudulent behaviours with external partners. Federated learning is capable of preserving customer privacy while enabling competing banks to share intelligence on bad actors, bolster machine learning models, and reduce operational costs.
Hosted by The Banker in partnership with Red Hat, this webinar convened leading technology and banking executives leveraging AI to fight financial crime. Chaired by Elizabeth Lumley, Deputy Editor of The Banker, our panel explored use cases for automation in fraud detection, while spotlighting advantages of federated learning for the payments industry and beyond.