United Airlines Bans AI Agents From Booking Flights

United Airlines has barred personal AI assistants from booking tickets on its flights, citing privacy exposure and federal regulatory requirements as the two decisive reasons. The airline set out its position in a policy page on its digital trust center, and the move was first reported by The Wall Street Journal.

The decision arrives despite minimal observed demand. United says it has not yet seen many people try to use AI agents to book flights, an admission that places the ban in the category of precautionary policy rather than reactive enforcement. The company's experience matches what other consumer businesses have documented: DoorDash, the food delivery platform, reported that it had received only several hundred orders from AI agents — a rounding error against the millions of orders placed through conventional channels in the same period.

What are United's stated concerns?

The airline identifies two distinct risk categories. The first is privacy: an autonomous agent acting on a customer's behalf could expose personal information that the customer never intended to share with a third-party system, or complete purchases the customer did not actually authorize. An AI assistant that misreads a fare, a date, or a passenger name could commit the traveler to a transaction with no straightforward path to reversal.

The second category is regulatory. United points to federal rules governing the information airlines must share with customers before and during a booking. Among the obligations the airline cites is the requirement that passengers attest their bags do not contain explosives — a security attestation that presupposes a human traveler making a legal declaration. An automated agent intermediating the purchase raises the question of who, exactly, is making that attestation, and whether a software system can lawfully make it on a person's behalf. For an airline, the compliance exposure of guessing wrong on that question outweighs the convenience of serving an early-adopter minority measured in hundreds of users industry-wide.

What does the ban change in practice?

The restriction is narrower than a wholesale rejection of AI tools. Agents can still display United flights to users — search, comparison, and itinerary research remain open to automated assistants. The boundary sits at the transaction itself: the moment money changes hands and a contractual relationship forms between passenger and carrier, United requires a human in the loop.

That distinction matters for how the booking interface of the near future takes shape. It preserves the airline's direct relationship with the customer at the point of purchase, along with the disclosure obligations attached to it, while leaving the research and planning layers of the travel workflow available to AI tooling. Other carriers watching United's move now face the same structural choice: expose booking APIs to agents and absorb the attestation problem, or hold the transaction behind a human-verified interface.

The policy also lands at a moment when AI agents are expanding from answering questions to executing transactions across consumer services. United's position — ban the transaction, permit the browsing — offers an early template that regulated industries can copy while regulators and standards bodies work out whether autonomous purchasing requires its own compliance framework.