Burger King is rolling out an AI system that scores employees on how friendly they are. The company calls it coaching. Labor groups call it something else.
The fast food chain started piloting an AI assistant called Patty at roughly 500 US locations earlier this year. By the end of 2026, it plans to deploy Patty across all 7,000 North American restaurants. The system listens in on drive-thru interactions, tracks service speed, and evaluates how warm and welcoming employees are to customers.
"This is all meant to be a coaching tool," Thibault Roux, Chief Digital Officer at Restaurant Brands International, Burger King's parent company, said of the system. He added that the company worked with franchisees and customers to "define friendliness in measurable terms."
That phrase is doing a lot of heavy lifting. Defining friendliness in measurable terms means an algorithm is now deciding what counts as warm, what counts as cold, and feeding that back to managers as data. For hourly workers making $12 to $15 an hour, the distinction between "coaching" and "surveillance" can feel pretty thin.
Burger King isn't alone. Yum! Brands, which owns Taco Bell, Pizza Hut and KFC, launched its own AI platform called Byte and has already deployed it to more than 35,000 restaurants globally across its 63,000-plus locations. Yum partnered with Nvidia to build out voice AI agents for drive-thrus, piloting them at 500 locations this year.
Dane Mathews, Taco Bell's Global Chief Digital and Technology Officer, has been candid about where the tech stands. "Sometimes it lets me down, but sometimes it really surprises me," he said, adding that the company now recommends restaurants use voice AI only during certain hours and closely monitor it during others. That kind of honesty is notable because not every chain has been as lucky with the technology. McDonald's ended its own IBM-powered drive-thru AI test in July 2024 after three years without expanding it.
On the smaller end, Andy Brown, founder of the 14-location Andy's Pizza chain in the DC area, is using AI differently. Rather than scoring workers, he's building custom software to handle operational tasks. "As a non-technical individual with in-depth knowledge of my business, I can create software tailored to my operational needs," Brown told Axios. It's a reminder that "AI in restaurants" covers a wide range, from tools that help owners run their business to tools that watch employees do theirs.
The labor side of this is heating up. California State Senator Jerry McNerney introduced SB 947, the No Robo Bosses Act, which would require human oversight of AI workplace decisions. A predecessor bill passed both chambers last year before Governor Newsom vetoed it, calling the restrictions "overly broad."
Lorena Gonzalez, president of the California Federation of Labor Unions, put the current landscape bluntly: "Right now, there are absolutely no restrictions on how employers can use artificial intelligence to arbitrarily discipline and fire their workers."
That regulatory vacuum matters because the technology is scaling fast while the rules around it barely exist. A Deloitte survey found that 82% of restaurant executives plan to increase AI investments, but only 28% said they were prepared to manage risk and governance around those tools.
There's a version of this story where AI coaching tools genuinely help restaurant workers get better at their jobs and earn better tips. That version requires transparency, worker input, and actual limits on what the data can be used for. None of those things exist right now. What does exist is an industry racing to quantify every human interaction in a restaurant while the people being quantified have zero say in how it works. Burger King can call Patty a coaching tool all it wants, but if the person being coached can't see the scorecard, challenge the score, or opt out, then coaching is just a friendlier word for monitoring.
