Mastercard is leveraging artificial intelligence to enhance its fraud detection and prevention systems. The company's AI technology is designed to identify unusual patterns and block fraudulent transactions in real-time, significantly reducing false-positive fraud cases. This initiative is part of a broader effort to integrate AI innovations across various sectors of the company's operations.

For the past decade, Mastercard has been using machine learning algorithms to monitor transactions and detect anomalies such as multiple failed logins and large withdrawals. The newest AI-powered system scans approximately 160 billion transactions annually. Seckin Yilgoren, Mastercard's senior vice president of security solutions in North American markets, explained that the company processes vast amounts of data to identify patterns indicative of fraudulent activity.

Mastercard employs a risk-scoring system called Decision Intelligence, which assigns a score to each transaction based on hundreds of millions of data points, including the cardholder's name, address, and purchase history. Transactions falling below a certain threshold are flagged for potential fraud. This process occurs in under 50 milliseconds, enabling near real-time detection and prevention. Decision Intelligence continuously learns and adapts to new fraud patterns without manual intervention, reducing false positives by distinguishing between legitimate unusual activity and actual fraud.

In 2024, Mastercard introduced Decision Intelligence Pro, which assesses relationships between user characteristics and behaviors to validate transactions. Daryl Lim, an affiliate at the Center for Socially Responsible Artificial Intelligence, noted that while AI is compelling for fraud detection, it has limitations. Machine learning models can inadvertently associate legitimate transactions with fraud based on biased historical data, potentially disproportionately affecting certain demographics or locations. Lim advocates for a hybrid model combining AI's speed and scale with human judgment for nuanced insights.

Andrew Reiskind, Mastercard's chief data officer, emphasized the company's AI governance program, which includes human oversight to ensure ethical and responsible AI use. Mastercard also uses AI to detect fraud by mapping connections between accounts, devices, and transactions, employing behavioral biometrics to identify imposters. This involves analyzing how users type and swipe on apps, considering factors like typing cadence, device handling, and mouse movements.

To combat first-party fraud, where customers falsely request chargebacks, Mastercard has implemented the First-Party Trust program. This program uses AI to scan data points such as IP address, device ID, and shipping address to assess the legitimacy of chargeback requests. In 2024, Mastercard also launched Scam Protect, a suite of AI-powered solutions designed to prevent online scams. These initiatives aim to ensure data security and trust for customers, banks, and merchants alike.