United Airlines moves to block unauthorized AI agents from booking flights

United Airlines is drawing a new boundary around who can act on a traveler’s behalf. On November 11, 2026, the carrier said it is taking steps to block unauthorized third-party personal AI agents from its systems—including access through travel agencies and aggregators. It cites risks to customer accounts and compliance with aviation rules.

The restriction follows the agent through the booking chain

The policy is not limited to an assistant trying to book directly on United’s website. United says it is restricting access both within its digital platforms and indirectly through travel agencies and aggregators. That makes the intermediary route part of the same restriction, rather than an alternative path for an unauthorized agent to reach the airline’s booking systems.

United is also updating its booking, ticketing and distribution policies. Its announcement describes work underway, not a completed technical rollout. The stated target is unauthorized outside personal assistants; the airline is not announcing that customers must stop using ordinary travel-booking services or manage every reservation exclusively through United.

Aviation analyst Henry Harteveldt told CBS News that United fares will remain bookable through aggregators, naming Kayak, Hopper and Google Flights. That distinction matters for travelers: restrictions on software acting for them do not, in his account, remove the familiar services they use to find and book fares.

Credentials give an assistant power beyond finding a fare

United’s security argument starts with account access. Sharing credentials with an outside AI assistant can expose sensitive information, the airline says. Once a tool can take actions within an account, the potential consequences extend beyond a mistaken recommendation: United warns of unintended purchases, account changes or spending of MileagePlus miles. It presents these as risks, not documented incidents behind the announcement.

The airline also raises a separate concern about automated check-in. It says AI actions could violate Federal Aviation Administration or Department of Transportation regulations. Its example is the human acknowledgment at check-in that bags contain no banned substances, such as explosives. This is United’s explanation of a potential regulatory conflict, not an announced finding by either federal agency.

For customers, United promises more guidance through an update to its Digital Trust Center. That update is intended to explain what the restrictions mean, how to protect an account and where to seek help if an agent takes an unintended action. The announcement does not specify an enforcement timetable or explain how unauthorized agents will be identified.

United is saying they don't want to add yet another, and potentially even more powerful, intermediary than they already face in terms of Google search, price-comparison metasearch sites and online travel agencies.

Henry Harteveldt, aviation analyst at Atmosphere Research Group, speaking to CBS News

Security is not the only stake in booking access

Harteveldt sees a commercial interest alongside the security case. He said the policy suggests United wants to retain control of the customer booking process. He also pointed to the cost of flight searches, arguing that the airline would want to avoid expenses associated with a surge of AI-bot searches. That is his assessment of the incentive, not United’s stated rationale.

His consumer assessment is narrower: travelers should not find it harder to locate good United fares through existing channels, but blocking bots will not make those fares easier to find either. United, meanwhile, says it will keep evaluating how personal AI assistants interact with its services as the tools evolve, with customer protection as its focus.