McDonald's AI Tells Franchisees What to Charge for a Big Mac

Here's a thought experiment that isn't one: the price of your Big Mac gets set by an algorithm that has already worked out exactly what your zip code will stomach. That, at least, is what a Reuters investigation found inside McDonald's pricing machine — a machine-learning engine chewing through millions of daily transactions across nearly 14,000 American restaurants, spitting out what the company calls "the optimal price" for every menu item at every location.

McDonald's insists none of this is happening the way it sounds. "AI does not set the price of a Big Mac or any other menu item," the company says. Its franchisees tell a different story. "You don't really have much of a choice anymore," said Karen King, a former store owner who retired from the franchise system last summer. Somebody is not telling the truth about the most eaten burger in America.

Photo: A Big Mac, photographed in Japan — illustrative, not one of the Fresno stores. The Reuters probe found two company-run locations two miles apart in Fresno charging $5.69 and $6.89 for the identical burger. (Wikimedia Commons, freely licensed.)

Inside the pricing engine

Reuters' piece — out September 29, by Waylon Cunningham — leans on access you don't usually get: August screenshots of the company's own pricing portal, plus nine people who have seen the chain's strategy up close. What they describe is not a spreadsheet. It is an AI platform, run for McDonald's by a vendor called Tiger Analytics (which declined to comment), that has been operating in some form since at least 2019.

The engine's job is relentless optimization. Machine-learning algorithms continually analyze data from millions of daily purchases and generate location-specific price recommendations — the "optimal price" for a Big Mac in Fresno versus a Big Mac in Fresno two miles away, down to discounted coffee for seniors. The portal even tells franchisees how price-sensitive their customers are. One screenshot reviewed by Reuters carried the message: "Your restaurant is showing MEDIUM SENSITIVITY to Price" — calculated, in part, from "customer willingness to pay in your area."

Read that twice. Not beef. Not rent, not wages, not the electric bill. Your willingness to pay — that is the input. On top of it, the platform scrapes the public online menus of nearby rivals — Wendy's and Burger King among them, and both told Reuters they don't do AI pricing themselves. (Anyone remember Wendy's 2024 meltdown? Its CEO floated "dynamic pricing," the internet lost it, and the company spent months swearing we'd all misheard.)

Two ex-Tiger staffers described the engine's rules of engagement, and honestly, they read like a guerrilla manual for squeezing margin without starting a riot: hike the items that haven't moved in two-plus years; only touch what's already gone up in at least 30 percent of stores; and leave ice cream and drinks alone all summer. There is even a section in the Reuters piece titled "HOLDING THE LINE ON SUMMER ICE CREAM PRICES," which would be funny if it weren't the product of a machine deciding what your McFlurry is worth in July.

The $5.69 Big Mac two miles from the $6.89 Big Mac

The numbers that made the story travel are from Fresno, California. Checking the McDonald's mobile app, Reuters found a company-run store selling a Big Mac for $5.69 — and another company-run store, two miles away, charging $6.89. A 21 percent premium. Same burger, same company, same city.

The New York Post went further, sending reporters to Fresno with cameras. At 4505 E. Cesar Chavez Boulevard: $5.69. At 4898 E. Kings Canyon Road, 1.3 miles away: $6.69 — a dollar more. The gaps extended across the menu: an M&M McFlurry at $4.69 versus $5.19, a Deluxe McCrispy at $5.99 versus $6.49.

We compared the two outlets' figures and they don't quite line up — Reuters' app-based $5.69/$6.89 pair and the Post's on-the-ground $5.69/$6.69 pair are different stores checked on different days. The gap is real; the exact pairing depends on when you look. That fluidity is itself the point: these prices move.

The customers the Post interviewed were blindsided. "I always thought they were all the same," one said. Leticia Oliva, told about the cheaper store down the road, said she didn't think the differences were fair — and that knowing about them would make her less likely to return to the pricier one. Gabriel Corral-Cardenas said he had no idea prices differed at all. There is something quietly devastating about that: millions of people buying the most standardized product on earth, assuming the standardization extends to the price.

To be fair — and Reuters is explicit about this — the investigation could not confirm that the pricing engine's recommendations caused any specific gap. McDonald's notes, correctly, that costs vary and that stores a few miles apart can belong to distinct markets. The Fresno numbers are suggestive, not dispositive. What moves the story from suggestive to damning is what the franchisees say happens when they ignore the machine.

"You don't really have much of a choice anymore"

McDonald's official line is that franchisees are free to set their own prices. Five store owners told Reuters the company pressured them to use the AI pricing tools anyway. King, the retired franchisee, described phone calls from corporate officers when owners strayed from the recommendations. One former franchisee said she didn't feel compelled — and five others described various forms of pressure. That is not a rogue anecdote; that is a pattern with a dissenter.

The paper trail backs them up. In January, McDonald's began requiring franchisees to be "constructively engaging with McDonald's approved Pricing Consultant and Tools" as part of new business standards, according to internal communications reviewed by Reuters. A franchisee document from June shows the company records deviations from the recommendations in detail, with guidance issued at least three times a year. On the company's August earnings call, CEO Chris Kempczinski publicly called out the roughly one-third of franchisees who hadn't complied with under-$3 menu pricing guidance, noting their "business results were a lot softer." Read that sentence again. That is the CEO of McDonald's telling investors that franchisees who disobey pricing guidance perform worse. The "tool, not a mandate" framing starts to wobble.

Then there's the $18 Big Mac meal. In 2023, the pricing tools suggested Connecticut franchisee George Michell charge about $18 for a Big Mac meal at his turnpike location. Michell later sued McDonald's in federal court in the Eastern District of New York (case 1:24-cv-03442), alleging the company tried to push him out of the system for discriminatory reasons; McDonald's says he repeatedly breached his franchise agreements, and the case is ongoing. A franchisee filing last December noted, dryly, that although the $18 price went viral and sparked nationwide outrage, it "caused no loss of sales." People paid it. The algorithm knew.

McDonald's: "AI does not set the price of a Big Mac"

The company's denial has been total, disciplined, and repeated to every outlet that asked. To Reuters: the pricing portal is "a tool, not a mandate, designed to provide restaurant-specific recommendations to help franchisees deliver value for customers and make informed business decisions." The reporting, McDonald's said, rests on "speculative and uninformed" claims that "attempt to recast a standard business practice as something controversial."

To Inc.: "AI does not set the price of a Big Mac or any other menu item." To PEOPLE: the same sentence, plus confirmation that the tool "does not set prices based on individual willingness to pay" and is not used "to charge customers different prices depending on how much they are prepared to pay." On September 30, McDonald's executive Jon Banner took the fight to X: "For the record: AI does not set Big Mac prices, make automatic price changes, or determine what individual customers pay. @Reuters got it wrong."

Update, Oct. 3: McDonald's has escalated its response since this story first moved. On Thursday the company posted a "Separating Fact from Fiction" chart on its website insisting "AI Does Not Set Prices," with a spokesman dismissing the complaints as "more sizzle than substance" — "This is a nothingburger with maybe extra pickles," he told the New York Post. "Dynamic pricing is not on the menu at McDonald's. It's never been on the menu." The fight comes as the chain launches its annual Monopoly promotion on October 6, with Wall Street analysts growing cautious on the stock (ad-hoc-news, Oct 3). The company line hasn't changed — but it's getting louder.

The company's substantive defense has two parts. First, the algorithm doesn't change prices in real time — it makes recommendations, periodically, which franchisees can accept or reject. Second, pricing-recommendation analytics are, as the spokesperson put it, "common across industries."

Both claims are narrowly true and broadly beside the point. Nobody alleged a live auction for your Quarter Pounder. And "everyone does it" is a strange defense for a company simultaneously arguing its tool is nothing special. The question was never whether the software literally edits the menu board. It's whether a "recommendation" you get phone calls about disobeying is a recommendation at all.

The antitrust problem

Here is where the story gets genuinely dangerous for McDonald's. The pricing portal's own legal terms — visible in the screenshots Reuters reviewed — warn franchisees that owners "may be competitors of each other" and that it is "particularly important for all Users of the Tool to understand and comply fully with anti-trust and competition laws."

William Kovacic, who directs the competition law center at George Washington University and served as an FTC commissioner, told Reuters that language is "an acknowledgment there's a potential problem." To Inc., he went further: "A crucial factor is whether the formal policy is undercut by day-to-day pressure to abide by [the suggested price]... I would expect that if there were a challenge, you'd see a lot of efforts to collect evidence that showed that, in practice, day in and day out, they apply pressure to their franchisees to behave in a way that is not reflected in that caution about obeying the law." He compared the setup to the RealPage and Agri Stats cases — the "hub-and-spoke" model in which one company allegedly orchestrates pricing across nominally independent clients through shared software — and called it "new legal territory."

Fiona Scott Morton, the Yale economist and former chief economist of the DOJ's Antitrust Division, told Inc. she finds it "very surprising that McDonald's would encourage franchisees to engage in a pricing practice at the same time that it warns them that it might be an antitrust violation" — suggesting the company is shifting legal exposure onto its franchisees, "a strategy she doubts will hold up."

The counterpoint comes from Jean-Pierre Dubé of the University of Chicago's Booth School: "Zone pricing is fairly prevalent in retailing and QSR more broadly... Unless a product's price is advertised nationally, or even regionally, it's not that uncommon for some chains to charge different prices in different locations." He cited a Chicago grocery chain running multiple price zones within a single county back in the 1980s — before any of this was algorithmic.

For now, the regulators are silent. The FTC and DOJ didn't respond to Reuters' requests for comment, and no lawmaker has publicly weighed in. But the FTC is already circling the biggest AI labs with document requests and sworn testimony — algorithmic pricing is unlikely to stay off its radar for long. But the advocacy world has its teeth out: antitrust lawyer Lee Hepner, writing in Matt Stoller's BIG newsletter, branded the whole apparatus "surveillance pricing," noting McDonald's bought the personalization firm Dynamic Yield back in 2019. Instacart, facing similar heat, killed an AI pricing test last December after a study drew fire from consumers and lawmakers. The pattern is forming, even if the case law hasn't.

McDonald's AI pricing: the questions everyone asks

Does AI set Big Mac prices?
McDonald's says no: the tool gives recommendations, franchisees decide. But five franchisees told Reuters the company pressured them to follow the AI's prices, and deviation is logged (Reuters, Sept 29).

What is Tiger Analytics?
The vendor running McDonald's pricing platform (declined to comment); the engine has operated in some form since at least 2019, analysing millions of daily transactions.

How big are the price gaps?
Reuters found a company-run Fresno store at $5.69 and another two miles away at $6.89 — a 21% premium for the same burger.

What's the antitrust angle?
The portal's own terms warn franchisees about antitrust law; ex-FTC commissioner William Kovacic compared the setup to the RealPage "hub-and-spoke" cases and called it "new legal territory" (Reuters).

How has McDonald's responded?
A "Separating Fact from Fiction" chart posted Oct 1 ("AI Does Not Set Prices"); exec Jon Banner's Sept 30 X post: "@Reuters got it wrong" (NY Post, Oct 1).

Our take

McDonald's wants this story to be about a misunderstanding: a helpful analytics tool, willfully misread by innuendo-hungry reporters. But the company's own documents describe a machine that scores your neighborhood's willingness to pay, tracks who disobeys it, and phones the disobedient — while its lawyers preemptively warn everyone involved about antitrust law. You don't put a seatbelt warning on a parked car.

The deeper dishonesty is the framing of "recommendations." A recommendation you can decline without consequence is advice. A recommendation that gets logged, reviewed three times a year, and punished on earnings calls is a quota with better PR. Five franchisees described the pressure; the January business-standards language made "constructive engagement" with the tools a requirement. This is what AI commercialization looks like up close — OpenAI is testing sponsored agents inside ChatGPT, and McDonald's has built a machine that prices your lunch by your zip code. At some point the distinction between the algorithm's suggestion and the company's instruction collapses — and McDonald's is asking us to believe it collapses nowhere.

And consider the timing. Placer.ai estimates show McDonald's US foot traffic declining year-over-year for every complete month since March. The response to fewer customers, apparently, is not lower prices — it's a smarter machine for finding the highest price each remaining customer will bear. Brooklyn customer Diane Bezucha, interviewed by Reuters, put the consumer's view plainly: it's in the company's interest to know where demand is, but if the technology is used to raise prices "simply because of more demand, that doesn't really help me as a customer."

She's right, and she's the whole story. The Big Mac was supposed to be the one price everyone knew. Now it's a different price for everyone, set by a machine, denied by a spokesperson, and enforced by a phone call. Seventy-three percent of Americans already say they don't trust AI companies. After this week, they might start adding fast-food chains to the list.

Sources: Reuters ("Inside McDonald's push to have AI price your Big Mac," Sept 29, 2026); Inc.; PEOPLE; New York Post/California Post; TheStreet; The Big Newsletter (Lee Hepner); Times Now.

By Daniel Rao, TechPulse USA

Daniel Rao covers AI, labor economics and US technology policy for TechPulse USA.