Financial institutions adopt AI to combat deepfakes and transaction fraud

Financial institutions are increasingly integrating artificial intelligence to combat evolving fraud threats. Ping An Digital Bank (International) Limited has completed the second cohort of the Hong Kong Monetary Authority’s Generative AI Sandbox. In partnership with OneConnect Financial Technology, the bank developed an ‘Anti-Fraud Strategy Platform’ designed to detect deepfake-related fraud. Testing of the platform, which uses AI facial recognition to verify user selfies in real time, demonstrated a detection accuracy exceeding 96%.

On a broader scale, a report by PYMNTS Intelligence and Visa DPS indicates that 89% of card issuers intend to adopt or expand AI fraud capabilities. Half of these issuers identify fraud detection and scoring as their most critical AI priority for the coming year. Additionally, 63% of issuers plan to implement real-time fraud detection to facilitate immediate transaction decisioning, aiming to prevent fraud without creating friction for legitimate cardholders.