United Airlines Loves AI. Just Not Yours.

I have to give United Airlines credit. It takes a special kind of confidence to spend years using technology to figure out how much money you can get out of customers, then turn around and tell those same customers they can't use AI to help book a flight.

Apparently, AI is great when it's working for United. When it's working for me, we have a problem.

United recently announced restrictions on third-party AI agents booking flights for customers. Their reasoning? Security, privacy, unauthorized transactions, and protecting customers.

How thoughtful.

I'm glad we're suddenly so concerned about protecting consumers. I was beginning to think the airline industry only cared about us when there was an opportunity to charge another $40.

And look, I understand the security argument. I don't particularly want an AI assistant booking me a surprise trip to Omaha because it misunderstood something I said.

But I also don't think it's particularly difficult to imagine a system where I authorize a purchase before the AI completes it. We manage to do that with plenty of other financial transactions.

So why is booking a flight suddenly such an impossible concept?

The part I find especially funny is that airlines have spent decades developing increasingly sophisticated pricing systems. They know how demand changes, when people are likely to book, which flights are filling up, and how to adjust prices to maximize revenue.

That's perfectly acceptable, apparently.

But now imagine I have an AI assistant that can do some of that work for me. It can look across airlines, compare prices, understand my preferences, and find the best option without me spending three hours opening 47 browser tabs.

Well, we certainly can't have that.

Even better, the FTC recently opened an inquiry into whether companies, including United, use personal information to determine individualized prices. That doesn't mean United is doing anything wrong, and demand-based pricing isn't the same thing as charging someone more based on their personal information.

But the timing is pretty funny.

The government is asking questions about how companies use information to determine what customers pay, and United is busy restricting how customers can use AI to make their own purchasing decisions.

I mean, come on.

And then there's United CEO Scott Kirby, who reportedly said, "To be honest, I don't want to create another middleman."

I actually appreciate that answer more than the security explanation.

At least it's direct.

Because I suspect that's a much bigger part of what this is really about. Airlines have spent years trying to keep customers inside their own websites and apps, where they control what we see, what gets promoted, and how we buy.

An AI assistant working for me doesn't particularly care what United wants me to see. It cares about what I asked it to find.

And I can see why that might be inconvenient.

What bothers me is the larger implication.

We're entering a world where companies can use AI to better understand our behavior, influence what we buy, and maximize what we spend. We're told this is progress, innovation, and the future of business.

But the moment consumers start using that same technology to make smarter decisions, suddenly there are all these concerns about whether we should be allowed to.

I'm not arguing that AI agents should have unlimited access to airline systems or customer accounts. Security matters. Consent matters. There are legitimate problems to solve.

I'm arguing that those problems should be solved, rather than becoming a convenient reason to keep the advantage on one side.

And maybe that's what irritates me most about this.

I actually think AI has enormous potential to make everyday life easier for people. Not just businesses. Not just the companies with the biggest technology budgets. Regular people who want to spend less time researching purchases, understand their options, and make better decisions.

That's the part of AI I'm excited about.

So it's frustrating to watch a company embrace the benefits of increasingly intelligent technology for itself while putting restrictions on customers who want to do the same.

United is perfectly happy to have technology help it figure out what I should pay.

I'd just like technology to help me decide whether I should pay it.

Apparently, that's asking a little too much.

Previous
Previous

200 Hours, Two Alphabets, and a Questionable Amount of Free Time

Next
Next

The Robot Monk and the Illusion of Care