Travelport, Booked AI, Basis Theory, and Mastercard have completed a live, end-to-end flight booking and payment carried out by an AI agent. Booked AI's travel agent searched for, booked, paid for, and ticketed a flight on a major international airline without the customer leaving the conversation, and without the agent seeing the underlying card number. The transaction ran on existing New Distribution Capability (NDC) and payment infrastructure.
How the agentic booking worked
Booked AI's agent sourced a live NDC offer from the airline through Travelport TripServices. Once the customer confirmed the flight and fare in the conversation, Mastercard Agent Pay enabled the agent-initiated payment using a tokenised credential, while the underlying card details were held in Basis Theory's PCI-compliant vault. After authorisation, the booking was processed and ticketed through Travelport's and the airline's own systems.
The customer approved the specific booking and amount before payment, and Mastercard's Verifiable Intent linked that approval to the transaction so the payment could be checked against the customer's instructions. Had the details not matched the approved intent, the payment would not have gone ahead, and any further purchase would have required new approval from the cardholder. The airline remained the merchant throughout, processing the payment without an intermediary checkout.
Airlines keep control of the sale
Andrew Jordan, Chief Technology Officer at Travelport, said AI agents will become a major channel for air travel, and that airlines should not have to give up control of the sale or the payment to take part. Eimear Creaven, President, Global Partnerships at Mastercard, said connecting AI experiences with established travel and payments infrastructure can speed up adoption while maintaining the security and control that travellers and travel businesses expect.
Mennan Yelkenci, Founder and CEO of Booked AI, said the customer approves once and that approval is what the payment is checked against, as the company aims to replicate a human travel agent from search to ticketing. James Armstead, CTO at Basis Theory, said structured use cases such as travel booking, where the agent can pay securely on the consumer's behalf, are likely to drive consumer use of agents.
