Mastercard rolled out an expanded AI governance framework to manage how artificial intelligence is developed, tested, and used across its core financial systems. The company said the framework is meant to provide consistent oversight as AI becomes more deeply embedded in global payment and security operations.

According to Mastercard, the framework applies across the company and covers AI used in fraud detection, credit and risk scoring, cybersecurity, customer authentication, and payment network analytics. It sets out requirements for documenting AI models, testing them before deployment, reviewing risks, and approving them for use in live systems.

Mastercard said it updated its governance approach in response to the growing use of generative AI and increased regulatory expectations from banks and other financial institutions that rely on its technology. The company said the framework is designed to support audits and provide consistent checks on higher-impact AI models.

The governance structure defines clear stages for an AI system’s life cycle, from development through deployment and ongoing monitoring. It includes required testing, assessments of how models reach decisions, and defined responsibilities across engineering, risk, compliance, and product teams. Mastercard said the framework applies to both the models it builds internally and the third-party AI components used in its services.

The company said the framework is intended to support customers in regulated industries, including banks, fintech firms, and digital commerce platforms. Mastercard added that as AI plays a larger role in processing transactions and preventing fraud, consistent governance controls are needed to maintain trust across its global payments network.