Skip to main content
Beyond Chat is an enterprise chat infrastructure for managing AI-powered communication, information access, and business processes from one centralized platform. It goes beyond the single-bot approach: multiple digital assistants can operate independently within the same organization, using different models, different documents, and different access permissions.

​
Chat configuration

Each assistant is defined as an independent configuration unit. Different business units — HR, IT support, operations, legal, accounting, risk — can create an assistant tailored to their needs. Per configuration you define the model and provider, language setting, system prompt, document sources, file-upload capabilities, and access permissions, so each assistant has a personalized knowledge base and behavior.

​
Isolated knowledge (RAG)

The Pre-Trained Dataset section is the RAG data layer where each assistant’s persistent institutional documents (PDF, Word, TXT, and others) are uploaded, indexed, and turned into its knowledge base. Because each assistant’s dataset is completely isolated from the others, the correct assistant accesses only the correct documents — the HR assistant knows only HR policies, the IT assistant works only with technical runbooks.

​
Model choice

The LLM Provider tab determines which model each assistant runs on. Beyond Chat isn’t tied to a single model; cloud-based and on-premise models can be used together in the same environment, so you choose the most suitable model per assistant for both performance and security.

​
Instant file upload

Instant File Upload lets employees upload files directly in the chat and have analysis performed on them. These files are not added to the permanent dataset — they’re used only for the duration of the session. Per configuration you set the accepted file types, maximum size, number of uploads, and the model parameters used for file analysis. This is especially valuable for contract review, report analysis, technical log inspection, and instant document explanation.

​
Access control

The Permission & AD Configuration layer determines which users or groups can access each assistant. You can make an assistant open to everyone or establish group-based access through Active Directory — so only HR teams reach the HR assistant, only IT reaches the runbook assistant, and everyone can reach a general FAQ bot. Combined with multi-tenant support, internal teams, PoC environments, and partners can all be managed under one system.

​
Security behind every message

In the chat interface, security controls run entirely in the background — the user experiences a natural conversation. Clicking the info panel on a response reveals the security analysis: every active security header applied to that message, each control’s execution status, and its SAFE / UNSAFE result. Users can transparently see which security layers a response passed through. The Chat Detail screen shows a session’s message flow, sender information, timestamps, and the model’s token statistics, while Chat History lets conversations be reviewed at organizational scale with filters by date, user, organization, or assistant — meeting both usage-analysis and audit needs.

Model Library

The models each assistant can run on.

Data Protection

PII/PHI/CII controls applied to chat.

Logs & Telemetry

Chat interactions in the audit log.

Platform Overview

Where Beyond Chat sits in the platform.