AI in the Payment Industry by Karamjit Singh (Mastercard)
- Predictive AI
Overview
In this talk from an H2O.ai event, Karamjit Singh, Director of AI Products at Mastercard, walks through how AI is applied across the payments industry. He covers fraud detection, including third-party fraud like account takeover and synthetic identity, and first-party fraud like friendly fraud and chargebacks, how Mastercard's Decision Intelligence, Safety Net, and account-level scoring products assess risk in real time, and how AI is used to optimize transaction processing speed, customer experience, and interchange fee routing.
Key takeaways
- The difference between third-party fraud, such as account takeover, stolen and synthetic identity, and loan stacking, and first-party fraud, such as fronting, friendly fraud, chargebacks, and sleeper fraud
- How Mastercard's Decision Intelligence scores every transaction for fraud risk in real time, at the moment of purchase
- How Safety Net detects anomalies at the merchant or terminal level to catch attacks as they happen
- How Account Intelligence generates predictive scores on a card, such as likelihood of compromise or an upcoming chargeback, so issuers can act before fraud occurs
- What actually happens behind the scenes during a card swipe, from merchant to acquirer to network to issuer and back, in the two seconds before approval
- How AI can optimize interchange fee routing and other decisions that are still largely rule based today
Start owning your AI