> For the complete documentation index, see [llms.txt](https://docs.guestway.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.guestway.io/ai/ai-center/about-how-the-ai-answers-guests.md).

# About how the AI answers guests

This page explains why the AI answers guests the way it does, and why the type and scope of your knowledge matter.

## What the AI knows about a stay

When a guest writes, the AI can draft a reply for your team, and Auto-Pilot can send it. The quality of the reply depends on what the AI knows. It combines two sources: your knowledge base in the AI center, and the details of the reservation and the property, such as the arrival and departure dates, the check-in and check-out times, the listing, its amenities, custom fields and your branding.

The AI never makes up property-specific facts. If the answer is in neither source, it says it will check and follow up, or it leaves the conversation for your team. That is why a missing answer is usually a knowledge gap, not an AI error, and why the listing details in **Properties** matter as much as the AI center.

## The four types of knowledge

The types differ in how much freedom the AI has with your words.

* **A Q\&A** is the truth for one topic. The AI keeps every relevant detail you wrote, rather than shortening it, and translates it into the reply language when needed. That makes Q\&As the best fit for short facts that guests ask about again and again.
* **A saved reply** is used almost word for word. The AI fills in the placeholders and adjusts the greeting, but does not rewrite it. Use it when the exact wording matters.
* **A guest guidebook** is background. The AI picks out only the part that answers the question. Smaller, focused documents give better answers than one large file.
* **An internal document** shapes the answer without being shown. Guests never see the document, but the AI reads it. That makes it the place for procedures such as escalation rules or late arrivals, and the wrong place for anything a guest must never learn, such as staff-only codes or variable pricing.

## How the AI learns from your team

Your team already answers guests every day. When it sends something noticeably different from the AI's draft, Guestway looks for a reusable fact in that reply and proposes it as a suggestion. One-off answers and personal details are filtered out. An empty suggestion list usually means that the AI's drafts already match how your team answers.

A suggestion that conflicts with an existing entry never overwrites it. Guestway cannot know which of the two answers is right, so the decision stays with you.

## Why Auto-Pilot leaves whole messages

Guests often ask several things in one message. If even one topic in the message is not allowed, Auto-Pilot leaves the whole message for your team. Half an answer from the AI, and the other half from your team later, would confuse the guest more than one complete answer a little later.

## Why an old answer can stay

After you edit a guidebook or saved reply, the change takes a moment to process. Until then, the AI still uses the old version. When the old answer stays after processing, the information usually lives in a second place, such as another Q\&A or an older guidebook. The AI cannot tell which one is current. For the same reason, the counters of an entry start again from zero when you edit it: the numbers always describe the current version.

## Related

* [AI center](https://docs.guestway.io/ai/ai-center.md)
* [How to add knowledge to the AI](https://docs.guestway.io/ai/ai-center/how-to-add-knowledge-to-the-ai.md)
* [How to set up Auto-Pilot](https://docs.guestway.io/ai/ai-center/how-to-set-up-auto-pilot.md)


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.guestway.io/ai/ai-center/about-how-the-ai-answers-guests.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
