# United Airlines wants you to check in, not your chatbot > Source: > Published: 2026-11-12 15:37:00+00:00 # United Airlines wants you to check in, not your chatbot [The Register](https://www.theregister.com) AI agents face restrictions over safety checks that require a human touch United Airlines has become the latest company to slap restrictions on unauthorized third-party AI agents, warning that the bots could make unintended purchases, expose customer data, or bypass checks requiring human acknowledgment. The airline [announced](https://www.united.com/en/us/newsroom/announcements/cision-125496) restrictions covering both its own digital platforms and indirect access through travel agencies and aggregators, with changes to its booking, ticketing, and distribution policies. United says automated actions could violate aviation regulations, citing the requirement for a human acknowledgment during check-in that bags contain no banned substances. A bot cheerfully ticking the box isn't quite the passenger confirmation it has in mind. And then there is the worry that "sharing account credentials with personal, third-party, AI assistants can expose sensitive information, and actions taken by these tools may result in unintended purchases, account changes, or use of MileagePlus miles." United would also rather customers use its own app than turn third-party assistants loose on its systems. AI agents are pitched as a way to outsource life's minutiae, and it's difficult to navigate the web without tripping over a breathless influencer explaining why you need one, or dozens, or hundreds of virtual personal assistants. The thing is that operating these assistants requires care and thought. Earlier this year, British mathematician Professor Hannah Fry [conducted a cautionary experiment](https://www.theregister.com/software/2026/05/05/british-mathematician-hands-openclaw-agent-a-credit-card/5228654) in which she gave an [AI agent](https://www.machinebrief.com/glossary/ai-agent) a bank card, and a bit too much agency, to see what it would do. Passwords were coughed up, and some ill-considered purchases were made. Still, as Fry said of the agent, dubbed Cassandra, "don't let her incompetence fool you, because these things are getting better fast." Last month, [reports](https://www.reuters.com/legal/government/us-banks-detect-unauthorized-ai-agents-customer-accounts-axios-reports-2026-10-22/) emerged of major US banks blocking third-party AI agents from operating inside accounts and taking actions on behalf of the customer. This month, several UK banks were [reported](https://www.telegraph.co.uk/business/2026/11/08/banks-ban-ai-from-accounts-over-fears-of-rogue-agents/) to have followed suit. United stated: "Corporate leaders in the banking and retail industries have taken similar actions regarding AI agents and we encourage others to consider measures that preserve the data and security of their customers and strengthen safety." The Register asked United Airlines how it planned to implement the blocking. There are a few options, including physical biometric confirmation for actions or checking for signs of automated browser driving. Amazon has also moved against unauthorized AI agents, blocking Meta's Muse (the "AI that gets it done," according to Meta's marketing) [in September](https://www.theregister.com/ai-and-ml/2026/09/21/amazon-shows-metas-muse-ai-shopping-agent-the-door/5297777). The airline has yet to respond. ® Get AI news in your inbox Daily digest of what matters in AI.