Why a message was stopped in the chat

Last updated: 11 October 2026

Before your message reaches the digital waiter, it goes through a few checks. When a check stops it, the chat shows a notice with a warning triangle that says so. This page explains what is checked, why, and what you can do.

1. What is checked

2. Why the safeguards are broad

The safeguards are broad on purpose. Stopping an ordinary message now and then is better than letting the digital waiter approve a discount, change a price or give allergen information that did not come from the restaurant's menu. So a perfectly innocent message can sometimes be stopped too, and we keep tuning them. They reduce the risk and do not remove it: with an allergy, or in any doubt, always check with the staff.

3. What to do when a message was stopped

4. A limit on the conversation

Each table has a limit on its conversation with the digital waiter, so the service stays quick and affordable for every restaurant. When you reach it, a waiter is called to your table and carries on with you, and you can keep ordering from the menu as usual.

5. What is kept

A stopped message is recorded in a block log: when, at which table, which check stopped it, and up to 300 characters of what was written, with no name and no account details. The log is used to see whether the safeguards stop too much, and it is deleted after 90 days, as set out in keeping and deleting data (Hebrew).

6. Stopped by mistake?

We would like to know. "Something not working?" at the bottom of the chat sends us a report, and you can tick the box to attach the conversation. For what the digital waiter receives and what it will not do, see use of artificial intelligence.