Workers say customers now trust AI hallucinations over their expertise
Across US restaurants, national parks and childcare settings, workers report customers increasingly believing AI-generated misinformation over professional advice, creating real safety risks.
The widespread use of AI chatbots like ChatGPT is reshaping customer service, as more people expect workers to go along with information the AI made up rather than trust professional expertise.
A New York City server, who asked to remain anonymous, says diners with food allergies sometimes dismiss her warnings about allergens in dishes, insisting ChatGPT told them otherwise — a dangerous gap since the dishes in question actually contained the allergens at issue. She's also seen growing numbers of guests reject the restaurant sommelier's guidance in favor of AI wine recommendations, sometimes requesting bottles that either aren't on the menu or don't exist at all, products of AI "hallucinations."
A pattern across industries
Similar experiences have surfaced among movie theater staff, baristas, and Apple Store employees, who've fielded requests for products or features that simply don't exist. A federal river ranger working in the Mountain West says visitor groups increasingly arrive with AI-generated itineraries containing geographically impossible plans, such as routes requiring boats to travel upstream. Even when rangers point to physical maps to explain the error, some groups remain skeptical, trusting the AI over rangers, maps, and guidebooks.
An after-school program teacher in New York reports that parents and caregivers increasingly consult ChatGPT or Claude about childhood development questions, sometimes overriding professional advice. In one case, a family insisted their AI-generated sleep schedule was fine for their child despite the teacher's assessment that the child needed significantly more sleep based on his behavior.
A food service worker with 15 years of experience says customers now commonly photograph restaurant menus and ask AI chatbots for dish or wine pairing recommendations instead of consulting staff directly. Her restaurant has also received reservation calls and emails that appeared to be generated by AI agents acting on behalf of customers rather than the customers themselves, blurring the line between human and automated communication.
