---
title: "AI Chat with Company Knowledge: an Assistant That Knows Your Company"
description: "Why an AI chat only becomes truly useful with your company data – and how an assistant with context works concretely in everyday life."
type: "wissen"
product: "assist"
slug: "ai-chat-with-company-knowledge"
source_language: "de"
target_languages: ["de", "en", "es", "pl", "tr"]
published: "2026-06-10"
status: "publish"
faq_json: [{"q":"What is an AI chat with company knowledge?","a":"An AI assistant that is connected to your company's real data – contacts, tasks, deadlines, notes. It answers based on your context instead of general guesses."}, {"q":"How does it differ from a normal chatbot?","a":"An ordinary chatbot starts from scratch and doesn't know your company. A chat with company knowledge accesses your data base and can store results directly there – for example creating a task instead of just producing text."}, {"q":"Do I have to explain everything to the AI again every time?","a":"No – that's the point. If the knowledge is structured within a platform, it's available to the assistant in new conversations without you repeating it."}, {"q":"What happens to the results from the chat?","a":"Good systems store working results permanently – with webRichtung assist as artifacts with title, type and tags, so nothing gets lost in the chat history."}, {"q":"Is such a chat possible in a GDPR-compliant way?","a":"Yes, if the provider takes data protection seriously. webRichtung is developed and operated in Germany; your data remains yours."}]
language: "en"
source_id: "wissen/ki-chat-mit-firmenwissen"
source_hash: "3a2e760052fb743100d0bf139d84144812aaa2cc3a368b24b90dd9285c819813"
---

An AI chat with company knowledge is an assistant that doesn't start from scratch, but is connected to your company's real data: contacts, tasks, deadlines, notes, documents. It answers questions based on your context and can store results directly in your data base – instead of just producing text that you then transfer by hand.

## Why a Normal Chatbot Isn't Enough in Everyday Life

AI rarely fails because of intelligence, but because of a lack of context. A regular chat knows neither your customers nor your open tasks nor the way you write quotes. The result: you explain the same background anew in every conversation, copy results by hand into other tools – and get answers that remain generic.

If tasks are in one tool, contacts in a second and documents in a third, even the best AI can't help reliably. The key is a shared data base that the chat is allowed to access.

## What Changes When the Chat Knows Your Data

With company knowledge, a text generator becomes a working tool:

- **Answers with substance:** The assistant relies on your real data instead of guesswork.
- **Results without media breaks:** "Create a task for Friday for that" – and the task is in your data base, not just in the chat window.
- **Less repetition:** You don't have to explain what makes your company unique in every conversation.

[webRichtung assist](https://www.webrichtung.de/module/assist/) is built exactly this way: the chat is connected to the platform and creates tasks, deadlines and notes directly in your data base.

## How This Looks in Practice

A typical working day with a connected assistant:

1. You ask for the draft of a rejection to a prospect – with goal, context and desired form.
2. From within the conversation, a task with a due date arises in your workflow.
3. The result is saved as an **artifact** – with title, type and tags, filterable and reusable.
4. You check and reschedule appointments within the same dialogue, without switching applications.

This way no good result gets lost in the chat history, and thoughts become binding steps.

## Explain Recurring Things Once: Skills

A chat with company knowledge becomes truly powerful when you teach it *how* your company works. At assist this is called **Skills**: recurring processes, rules or task patterns – such as "This is how we write quotes" or "This is how we respond to complaints". Once stored as a Skill, the knowledge is available to you and, if desired, to your entire organization. More on this in the article [Skills for recurring tasks](/en/wissen/ki-skills-wiederkehrende-aufgaben.html).

## What to Look For When Choosing

- **Data base instead of isolated solution:** Can the chat really create tasks, deadlines and notes – or does it only produce text?
- **Results that stay:** Are working results saved and stored in a findable way?
- **Data protection:** Where is the data processed? webRichtung is developed and operated in Germany, GDPR-conscious.
- **Fair pricing model:** With webRichtung, the account and users are free (Pay per Use) – you only pay for usage.

How you work with the chat and what artifacts exactly are is shown in the documentation at [docs.webrichtung.de/assist/](https://docs.webrichtung.de/assist/).