Burger King Tests AI Headsets that Listen for “Please” — and Everything else

With input from the Hill and NBC News.

Burger King is quietly rolling out a pretty futuristic (and slightly creepy) tool: AI-powered headsets that help workers with recipes, ping managers when supplies run low — and even flag whether employees are saying “welcome,” “please” or “thank you.”

The pilot, now live in about 500 US locations, feeds audio and operational signals to a voice assistant the company calls Patty. Patty answers workers’ questions in real time — how to make a menu item, whether to pull a limited-time burger from the digital menu if the bun or sauce is gone — and speaks back through the headset. It also routes alerts to managers: a drink machine running dry, a customer QR-reported messy restroom, or inventory that needs attention.

The system sits inside a new platform, BK Assistant, that Restaurant Brands International says will be available to all its US restaurants later this year. The parent company behind the rollout said the headsets are powered by an OpenAI model, which helps Patty understand and respond to requests and spot operational issues quickly.

There’s no denying the upside: fewer dumb mistakes on new menu launches, faster restocking, and a hands-free way to walk a rookie worker through an order while keeping their eyes on the line. In tight labor markets, tools that shave seconds off every transaction can add up to real savings.

But the politeness-tracking feature is what’s getting people’s attention. The software logs keywords that it treats as signs of “friendliness” — things like welcome, please and thank you — and shares that data with store managers. Burger King is stressing that the feature is meant as a coaching tool, not a punch-clock for manners. In a company statement, executives said the aim is to “reinforce great hospitality” and give managers real-time insights to reward teams, not to score or script employees.

Thibault Roux, Burger King’s chief digital officer, has described the approach as a way to iterate on what “friendliness” means from an operational standpoint and to use that signal to coach staff. Still, critics — and some workers — see surveillance risks. Headsets that listen to interactions raise obvious privacy and labor questions: Who reviews the data? How long is it stored? Could “politeness metrics” be folded into performance reviews or wage decisions down the line?

This is hardly an isolated experiment. Other big chains are racing into AI for kitchens and stores: Yum Brands has teamed with Nvidia on generative-AI projects, and McDonald’s has pivoted from a canned IBM partnership to work with Google on AI systems. Fast food is fast to adopt anything that promises efficiency gains, but that speed often outpaces public debate about worker protections.

On the shop-floor level, the tech could be useful. A manager getting an immediate alert that a milkshake machine’s running low can prevent a five-minute outage that backs up a line. Patty can also remove sold-out items from the digital menu automatically — a tiny convenience with big customer-service dividends.

But the social debate is bigger. Labor advocates worry about the escalation from helpful assistant to performance policing. Even if the company insists the tool won’t “score individuals,” terms like that can be slippery in practice. And there’s the optics problem: a brand preaching warmth while deploying an algorithm to monitor niceties feels tone-deaf to many.

For now, Burger King frames Patty as a colleague that helps people do their jobs better. For employees and customers, the question is whether that “help” will stay helpful — and stay human.