Visa, Mastercard, Ant Launch AI Agent Trust Framework

Visa, Mastercard, and Ant International have launched a collaborative initiative to establish common standards for identifying and verifying autonomous AI agents that conduct purchases on behalf of users, addressing security challenges in automated commerce.

"Trust is the foundation of the AI transformation," said Jiang-Ming Yang, chief innovation officer of Ant International, noting that AI agents' potential for hallucination means "we need to make sure people feel safe."

The Know-Your-Agent Interoperability Framework

The initiative centers on a joint Know-Your-Agent interoperability framework. According to an AOL report, the architecture is designed to allow traditional card networks, digital wallets, agent platforms, and online marketplaces to recognize trusted AI systems across fragmented payment ecosystems.

This cooperative effort builds directly upon existing proprietary technologies from the participating firms. Specifically, the framework integrates Visa's Trusted Agent Protocol, Mastercard Verifiable Intent, and Ant International's Agentic Mobile Protocol.

Platform Collaboration in Singapore

The three payment giants are conducting their joint development through BuildFin.ai, an innovation platform convened by the Monetary Authority of Singapore. By housing the project within BuildFin.ai, the participants aim to align regulatory expectations with technical standards for cross-border agentic transactions.

"Interoperability across Know-Your-Agent frameworks is essential to making agentic commerce work at scale," said Pablo Fourez, chief digital officer at Mastercard.

Projected Scale of Autonomous Commerce

The push for standardized verification protocols arrives as financial institutions anticipate a massive expansion in machine-driven transactions. According to a CNBC report, the companies cited projections from McKinsey & Company estimating that AI agents will handle between $3 trillion and $5 trillion of global consumer commerce by 2030.

What Is at Stake for Digital Payments

As autonomous software agents transition from retrieving information to executing financial transactions, legacy payment rails face novel security vectors, including unauthorized agent instructions and model hallucinations. Creating a unified credentialing standard seeks to mitigate these operational risks before autonomous shopping scales into mainstream consumer markets.