How it works
The assistant turns a request such as 'book me a desk near my team on Thursday' into actions in the workplace system: it checks availability, applies booking rules, picks a desk and confirms. Follow-up messages narrow the request, for example a room for six with video at ten.
It lives where people already work: Microsoft Teams, Slack, a mobile app, or a general AI assistant connected through the Model Context Protocol. That matters more than the language model, because adoption depends on not asking people to open another tool.
The assistant acts with the employee's own permissions and the organization's booking rules, so it cannot book spaces the person could not book themselves. Good assistants say what they did, and the booking appears in the same system, calendar and reports as any other.
How to measure it
- Share of bookings made through the assistant, by team.
- Time to complete a booking compared with the app.
- Requests the assistant could not complete, reviewed monthly to find gaps.
Examples
- An employee in Slack types 'desk near the product team Wednesday' and gets a confirmed desk two seats from a teammate.
- A manager in Microsoft Teams books a room for five and desks for a visiting team in one request.
Common mistakes
- Launching a separate AI app instead of meeting people in Teams, Slack or the assistant they already use.
- Letting the assistant bypass booking rules, booking credits or neighborhood assignments.
- Measuring success by chat volume rather than by bookings completed and time saved.
How Kadence helps
Kadence AI books desks and rooms by chat inside Microsoft Teams and Slack, using the same booking rules and calendar sync as the rest of Kadence. For AI assistants such as Claude, ChatGPT or Copilot, the Kadence MCP server (early access, enabled by an admin) adds desk, room, parking and visitor booking and check-in with each user's existing permissions.
Kadence AI · Microsoft Teams · Slack
Frequently asked questions
What can an AI workplace assistant do?
Typical tasks are booking desks, rooms and parking, finding when teammates are in, inviting visitors and changing or cancelling bookings. The best assistants work inside chat tools people already use and follow the same rules as the booking system.
What is AI desk booking?
AI desk booking lets an employee book a desk by asking in plain language, for example for a desk near their team on a certain day. The assistant checks availability and booking rules, picks a suitable desk and confirms the booking.
Does an AI workplace assistant train on company data?
It depends on the vendor. Ask whether customer data is used for training, where data is processed, whether the assistant acts with each user's own permissions, and whether it can be switched off for teams that do not want it.
What is the difference between an AI workplace assistant and an MCP integration?
An AI workplace assistant is built into the workplace product or its chat apps. An MCP integration lets general assistants, such as Claude or Copilot, reach the same workplace tools, so people can book space from the assistant they already use for other work.
Related terms
- MCP (Model Context Protocol) An open standard that lets AI assistants use tools and data in other systems, such as booking a desk, without custom integrations.
- Agentic workplace A workplace where AI agents carry out tasks, such as booking space and drafting plans, with people approving decisions.
- Microsoft Teams and Slack booking Booking desks and rooms, seeing who is in and checking in from inside Microsoft Teams or Slack.
- Desk booking system Software that lets employees find and reserve a desk before they come in, check in on arrival and free desks nobody uses.
- Workspace booking Reserving any bookable space at work, such as desks, rooms, lockers, pods, parking or coworking space, from one system.
Kadence editorial team · Last reviewed