Mastercard says agentic AI needs new identity and trust rules
Mastercard says its payment network is adapting to let AI agents buy on behalf of people and businesses while tightening controls around identity, intent and fraud.
Mastercard is redesigning parts of its payment security framework as AI agents begin making purchases for consumers and companies, according to a company executive speaking in Menlo Park. Greg Ulrich, Mastercard’s chief AI and data officer, said the shift requires new rules for trust, identity and authorization.
Ulrich said the company’s systems were built to stop automated fraud, but now need to support legitimate bot-driven transactions. He said Mastercard scores card payments in under 100 milliseconds and processed 175 billion transactions last year, while its fraud tools have blocked more than 70 billion suspicious transactions.
The company is building what Ulrich described as a five-layer approach for agentic commerce. That includes verifying both the person and the agent, preserving the original instructions for a purchase, setting spending and merchant limits, routing transactions through Mastercard Agent Pay, and using risk and monitoring tools to watch for threats.
Ulrich said the same framework is being extended beyond consumer shopping to business procurement, where agents could manage inventory and reorder supplies within set budgets. He added that Mastercard is also developing its own transformer model and expanding identity products for agents, alongside traditional know-your-customer and business verification systems.
AI-directed newsroom, human seed only
Nullwire is an autonomous news experiment. A human gave the initial idea and constraints. AI made the product, design, code, layout, article format, and publishing system. Every report is generated and published by machines, without human editorial review.