Ant International, Visa and Mastercard Begin Shared Rules for AI Payment Agents

Ant International, Visa and Mastercard have started work on a shared framework for identifying AI agents that make payments. The immediate goal is to make an agent’s identity and trust signals usable across card networks, digital wallets, agent platforms and marketplaces without requiring a fresh registration at every stop.

The companies call the effort a Know-Your-Agent, or KYA, interoperability framework. Its design draws a line between common recognition and common control: participating systems would use shared principles to streamline onboarding and identification, while each network keeps its own verification and decisioning processes.

That distinction is central to the proposal. Ant International’s chief innovation officer, Jiang-Ming Yang, described the intended outcome this way: an agent registered with Ant would not need to register again with Visa or Mastercard. The framework is meant to reduce duplicate identity-verification work and integration complexity, not to replace the separate payment ecosystems.

The collaboration goes beyond a directory of agents. The companies plan to link each agent to a validated operator, cardholder, business or organization; assess it against security and behavioral requirements; and continuously monitor transaction-related and identity signals. Those measures are intended to establish accountability for an agent’s activity and support ongoing evaluations.

The proposed KYA checks

• Connect every agent to a validated person, business or other operator.
• Assess security and behavioral requirements before recognizing the agent.
• Use ongoing identity and transaction-related signals to monitor activity.

The three companies had each previously introduced their own approaches: Visa’s Trusted Agent Protocol, Mastercard Verifiable Intent and Ant International’s Agentic Mobile Protocol. This collaboration does not declare one of them the winner. Instead, it explores common principles that could make their trust signals recognizable across cards and wallets.

The companies will work through BuildFin.ai in Singapore on approaches to agent verification, accountability and risk management. The effort builds on the Safeguards for Agentic Finance at Runtime framework, which provides the stated foundation for this collaboration.

The companies cited a McKinsey projection that AI agents could orchestrate $3 trillion to $5 trillion of global consumer commerce by 2030. That estimate explains the urgency behind interoperable rules, but it is a forecast rather than evidence that this framework has been adopted or that it will resolve the risks of autonomous payment activity.