If you were hoping to let an AI assistant handle your holiday travel, United Airlines just said no. The airline announced Thursday that it’s restricting AI agents and autonomous assistants from booking flights and managing reservations on customers' behalf. That covers its own website and app, and it also reaches through travel agencies and aggregators that might pipe those agents in.
United’s reasoning comes in two parts. The first is security. When people hand their login details to a third-party AI assistant, the airline says that can expose sensitive information, and the bot’s actions could lead to purchases you didn’t mean to make, changes to your account, or MileagePlus miles getting spent. The second part is regulatory, and it’s more interesting. Certain steps in air travel are supposed to be done by an actual human, like confirming at check-in that your bag doesn’t contain banned items. An AI clicking “yes” for you muddies that.
The company says it’s updating its booking, ticketing and distribution policies to make the ban stick, though it didn’t say when it takes effect or how it’ll detect agents in practice. United isn’t alone here, either. Just last month, Bank of America urged customers not to share banking passwords with AI assistants. Big companies are clearly struggling to tell a real customer from software acting on that customer’s behalf.
This is a real collision. Tech firms are pitching “agentic AI” as the next big thing, a helper that books, pays and plans for you. But if airlines, banks and retailers start locking the doors, those agents won’t have much to do.
