What the FTC Investigation Into Delta and Uber Algorithmic Pricing Means for Your Wallet

What the FTC Investigation Into Delta and Uber Algorithmic Pricing Means for Your Wallet

You check a flight on Delta, close the tab, come back twenty minutes later, and the price jumped eighty bucks. Or you open Uber next to your friend at the exact same bar, requesting a ride to the exact same neighborhood, and your screen shows twelve dollars more than theirs. You aren't imagining things. And now, the Federal Trade Commission wants to see the code behind those price tags.

When the FTC ordered Delta Air Lines, Uber, and several other major companies to hand over internal data on their algorithmic pricing systems, it marked a turning point in how regulators look at everyday consumer prices. For years, companies brushed off price swings as simple supply and demand. Plenty of seats left? Cheap ticket. Heavy rain and a concert letting out? Surge pricing. Simple enough. Except modern pricing algorithms don't just look at inventory anymore. They look at you. Also making news in related news: The Structural Failure of Fragmented UK Energy Policy and Household Cost Mechanics.

That distinction is why federal investigators issued formal orders demanding the underlying data, software details, and internal communications from companies using automated pricing tools. The FTC isn't just auditing basic dynamic pricing. They're digging into surveillance pricing—the practice of using personal data, browsing habits, device types, location history, and behavioral profiles to guess the absolute highest price an individual customer is willing to pay.

Why Dynamic Pricing and Surveillance Pricing Are Two Different Beasts

Most people lump every automated price change into one bucket. That's a mistake, and companies love that confusion because it makes aggressive price discrimination sound like basic economics. Further information on this are explored by The Wall Street Journal.

Traditional dynamic pricing reacts to market conditions. Airlines have used yield management systems since the 1980s. If an Airbus A321 has 190 seats and 170 of them sell out three weeks before departure, the remaining twenty seats get more expensive. Every shopper looking at that flight at that exact second sees the same number. It's predictable, or at least tied to physical supply.

Surveillance pricing flips the equation. Instead of asking what a seat or a ride is worth based on availability, the algorithm asks what you are worth based on your digital footprint.

Think about the signals your phone broadcasts every single time you open an app or load a booking page. An algorithm can see whether you're searching from a brand-new iPhone Pro or a four-year-old budget Android. It knows if you're browsing from a wealthy zip code. It might infer whether you're booking a last-minute business trip on a corporate card versus comparing budget carriers for a family vacation six months out. Even your phone's battery percentage has been studied by researchers as a potential proxy for urgency.

If you're at 4% battery at midnight outside an airport terminal, how hard are you going to comparison shop? You aren't. You're going to tap "Confirm" on whatever Uber shows you.

When the FTC launched its inquiry into surveillance pricing—initially issuing orders to major intermediaries like Mastercard, Revionics, Bloomreach, and JPMorgan Chase before expanding scrutiny toward consumer-facing giants like Delta and Uber—regulators targeted the exact mechanics of these individualized price tags. They want to know what data feeds the models, who sells that data, and whether companies are quietly charging different people different prices for the exact same service at the exact same moment.

What Delta and Uber Actually Do With Pricing Algorithms

Neither Delta nor Uber operates in the dark ages of fixed price menus. Both companies rely heavily on real-time software to set fares, and both have drawn intense public scrutiny over how far those systems go.

Take Uber first. Ride-hailing popularized dynamic pricing through "surge" multipliers, which used to be transparent. You opened the app, saw a 2.5x lightning bolt, and knew prices were elevated because drivers were scarce. Over the last few years, Uber shifted to "upfront pricing." Instead of a flat formula based on miles, minutes, and a visible surge multiplier, Uber's algorithm calculates a custom fare before you book.

Behind the scenes, Uber's system weighs route demand, traffic patterns, driver supply, and historical rider behavior. Critics, drivers, and researchers have pointed out for years that upfront pricing breaks the link between what a rider pays and what a driver earns. Two riders standing ten feet apart can receive noticeably different upfront quotes for identical trips. Uber has repeatedly stated that it does not set prices based on an individual rider's personal demographic profile or phone battery level, pointing instead to hyper-local supply and demand shifts that change second by second. The FTC's data demands are designed to test those claims against actual internal records and model weights.

Delta's situation hits an even rawer nerve with travelers. Airlines already operate the most complex pricing grids in modern commerce, divided into dozens of fare buckets with baffling rules. Recently, Delta executives publicly discussed expanding the use of artificial intelligence in revenue management—working with pricing tech vendors to replace static rules with AI models that react faster and more granularly to demand signals.

When an airline uses AI to manage fares, the line between "predicting route demand" and "profiling the shopper" gets blurry fast. Even without logging into a SkyMiles account, tracking cookies, IP addresses, and search patterns tell an airline's booking engine plenty about your intent. If you search for a Tuesday-to-Thursday flight from Atlanta to New York three times in one morning, the system knows you probably have to take that trip. You aren't daydreaming about a beach vacation. You're going to a meeting, and someone—likely your employer—is going to pay for it.

You might wonder why this is even a federal investigation. Isn't a private business allowed to charge whatever it wants?

Mostly, yes. In the United States, charging different customers different prices isn't automatically illegal. Senior citizen discounts, student rates, and happy hour specials are all forms of price discrimination, and nobody calls the feds over a cheap matinee movie ticket.

The FTC is looking at three specific legal tripwires where algorithmic pricing crosses from clever business into unlawful conduct.

First is deception under Section 5 of the FTC Act. If a company tells consumers that prices reflect real-time supply and demand, or markets a loyalty program as giving members the "best available rate," while its algorithm secretly inflates prices for loyal customers who rarely price-shop elsewhere, that's a deceptive business practice. Loyalty penalties are real. Companies know that habitual customers stop checking competitors, which makes them prime targets for quiet algorithmic markups.

Second is unfair competition and algorithmic collusion. When multiple competitors rely on the same third-party pricing software or data brokers, algorithms can effectively coordinate price hikes without executives ever picking up the phone to fix prices. We've already seen the Department of Justice and FTC go after software platforms like RealPage in the rental housing market over this exact issue. When everyone's algorithm optimizes for maximum margin using shared data feeds, prices drift upward across the entire industry, and consumers lose the benefit of real competition.

Third is anti-discrimination law. Even if a company never feeds race, gender, or age into an algorithm, machine learning models are notorious for finding proxies. Zip codes, browser settings, shopping histories, and device data correlate heavily with protected characteristics. If an algorithm systematically quotes higher prices to shoppers in predominantly minority neighborhoods or low-income areas—a digital version of redlining—companies face massive legal exposure.

Common Myths About Algorithmic Pricing

Whenever news breaks about an FTC pricing probe, terrible advice floods social media. Half of it is outdated, and the other half actually backfires.

Myth 1: Incognito mode completely stops airlines from tracking you.
Opening a private browser window clears local cookies, which stops basic session tracking. It does not hide your IP address, your screen resolution, your operating system, your browser fingerprint, or your approximate location. More importantly, if you log into your Delta SkyMiles account or Uber profile to apply points or view saved payment methods, incognito mode does nothing at all. The company knows exactly who you are the second you sign in.

Myth 2: Prices always go up the closer you get to departure.
Usually true, not always. Algorithms constantly rebalance. If an airline's model overestimates business demand for a flight and realizes forty-eight hours before takeoff that twelve seats are still empty, prices can suddenly drop below what they cost two weeks earlier. Algorithms care about maximizing total revenue per flight, not sticking to a rigid calendar.

Myth 3: Companies need your name and Social Security number to profile you.
Identity isn't required for surveillance pricing. Data brokers build shadow profiles tied to device IDs and hashed email addresses. An algorithm doesn't care that you're John Smith from Chicago. It cares that Device_ID_88492 regularly buys coffee at high-end cafes, travels twice a month, uses a flagship smartphone, and rarely abandons a shopping cart once an item is added. That behavioral tag is more than enough to test a 15% price bump.

What You Can Actually Do to Beat Pricing Algorithms

You can't force Delta or Uber to hand over their source code, and federal investigations move slowly. Even if the FTC uncovers clear abuses, enforcement actions and new rules take months or years to change what shows up on your screen. In the meantime, you have to protect your own wallet.

Break your behavioral pattern before you buy.
Algorithms reward hesitation with higher prices when they detect high intent, but they also use discounts to recapture shoppers who look like they're walking away. If you're booking a flight, search for routes on an aggregator like Google Flights first without clicking through to the airline's site repeatedly. Once you see the baseline fare, go directly to the airline to book in a single clean session.

Compare logged-in versus logged-out prices.
Don't assume your loyalty status automatically gets you the lowest fare. Check flight prices on a separate device or browser without logging into your frequent flyer account first. Sometimes member-only sales save you money; other times, algorithms recognize a captive loyalty member and withhold the lowest promotional fare buckets that they show to price-sensitive newcomers.

Force competition on ride-hailing apps.
Never rely on a single ride-hailing app. Keep both Uber and Lyft installed, and check both before confirming a ride—especially at airports, hotels, and event venues. Even a thirty-second comparison frequently reveals a 15% to 30% gap between the two platforms for the exact same route. If both look absurdly high, walk two blocks away from the geofenced pickup zone (like an airport terminal or stadium exit) and request the ride again. Hyper-local surge zones often drop off sharply just outside the main perimeter.

Use a VPN strategically, not blindly.
Changing your virtual location via a VPN occasionally surfaces lower point-of-sale fares for international flights, particularly when booking foreign carriers that price tickets based on local purchasing power. For domestic U.S. flights and local ride-hails, a VPN rarely lowers prices and can sometimes trigger fraud-prevention blocks during checkout. Save it for international itineraries.

Audit your data broker footprint.
Surveillance pricing feeds on third-party data. Opt out of personalized advertising in your phone's privacy settings (Limit Ad Tracking on iOS, Delete Advertising ID on Android), clear stored site data regularly, and stop using your primary email address for every retail loyalty sign-up. The less cleanly an algorithm can stitch your browsing history together across apps, the harder it becomes to hit you with an individualized markup.

The FTC's orders to Delta, Uber, and their peers pull back the curtain on a pricing system that was never meant to be visible to regular buyers. Companies spent billions building software to figure out how much pain your budget can tolerate before you walk away. Until regulators draw hard lines around what personal data can legally touch a price tag, assume every screen you look at is testing your limit—and make sure you give the algorithm as little information to work with as possible.

MR

Miguel Rodriguez

Drawing on years of industry experience, Miguel Rodriguez provides thoughtful commentary and well-sourced reporting on the issues that shape our world.