--- title: 'Why your AI assistant lies about your menu — and how cite-or-die fixes it' description: 'Generic AI can invent prices, dishes, and hours. Learn why menu hallucination is structural and how source checks can contain the risk.' url: 'https://www.semperi.com/blog/why-ai-assistants-lie-about-your-menu' --- # Why your AI assistant lies about your menu — and how cite-or-die fixes it > Generic AI can invent prices, dishes, and hours. Learn why menu hallucination is structural and how source checks can contain the risk. *Updated: Sep 16, 2026* *May 12, 2026 · 9 min · AI Operations · Semperi Team, Restaurant growth research* Consider a common failure case: an AI assistant writes a warm reply to a complaint about a schnitzel, then invents a free dish or price that never existed on the menu. The prose sounds plausible, but the restaurant would be publishing a false promise. A source check should prevent it. This isn't a rare glitch. It's the default behavior of every general-purpose language model, and if you're using ChatGPT or Claude as a generic assistant for your restaurant, it is happening to you whether you've noticed or not. This post explains why the lying is structural — not a bug that gets patched next quarter — and what an architecture that actually prevents it looks like. ## Why do AI assistants make up menu items and prices? Language models are prediction engines. When you ask one to write a review reply mentioning your margherita pizza, it predicts the most plausible-sounding continuation. If it doesn't know your actual price, it doesn't say 'I don't know.' It generates a plausible price — €11.50, $13.95, whatever fits the genre of 'pizzeria reply.' The model is not lying in the human sense; it has no concept of your menu as a fact to be checked. Plausibility and truth are the same thing to it. Pasting your menu into the prompt helps, but it doesn't fix the problem. The model can still blend your menu with its training data, quote last season's price after you've updated it, or confidently describe a dish you 86'd two weeks ago. And nothing in the loop checks. The text just flows out, fluent and wrong. ## What does a menu hallucination actually cost a restaurant? - A reply quoting the wrong price becomes a promise. The customer screenshots it and asks you to honor it. Now you're issuing a discount you never offered — or arguing publicly with a customer holding receipts. - A made-up dish in a public reply tells every reader that the owner doesn't know their own menu. Trust damage compounds: people notice, and they remember. - Wrong hours in an AI-written post sends people to a dark dining room. Those become one-star reviews that say 'showed up, closed' — the hardest kind to recover from. - Allergy and ingredient errors are the nightmare case. An AI that improvises 'gluten-free' onto a dish that isn't is a liability event, not a typo. One hallucinated public reply can undo months of careful review management. That's why we treat hallucination as a safety problem, not a quality problem. ## What is cite-or-die and how does it work? A good review reply starts with the original review and the facts that matter. Semperi keeps those beside a prepared reply so the owner can check it before it goes live. Here's the part that matters: an output that asserts a fact — a price, a number, a dish name — without a citation is rejected automatically, before it reaches you and long before it reaches a customer. Not flagged. Not warned. Rejected. The guardrail isn't a suggestion in a prompt; it's a checker that runs on every output, and there is no setting to turn it off. We call it cite-or-die: no citation, no claim. > A generic assistant generates the most plausible sentence. A cited system generates the most plausible sentence that survives a fact check — and throws away the ones that don't. ## Can't I just tell ChatGPT to only use my menu? You can, and it will mostly comply, and 'mostly' is exactly the problem. Prompt instructions are requests, not constraints. Under a long conversation, a weird review, or an updated menu the model hasn't seen, the request gets ignored — silently. There's no rejection step, no citation trail, no way to audit after the fact which facts came from where. The difference between 'asked nicely' and 'architecturally enforced' is the difference between a sign that says 'please wash hands' and a tap that won't dispense food until you do. Cite-or-die also changes what happens when the data is missing. If Grace can't find a claim to support a sentence, she doesn't improvise — the version either drops the claim or shows the gap. 'I don't know' is a feature. Generic assistants don't have it. ## What this looks like day to day Semperi keeps the details behind prepared work visible and makes missing information clear. That gives the owner a better basis for the next decision. Semperi's trust page explains the product policy for sources, work states, owner authority, and recorded results. Workflow enforcement still depends on the specific connected path. To see what the free scan finds from available public sources, start at app.semperi.com/scan. --- AI teammates for local food, retail, and service businesses. Start with a free scan and see what needs attention. - Full site map for agents: https://www.semperi.com/llms.txt - Free business scan: https://app.semperi.com/scan - This page for humans: https://www.semperi.com/blog/why-ai-assistants-lie-about-your-menu