Skip to main content
Microsoft
separator
https://catalogartifact.azureedge.net/publicartifacts/jespers_forretning.ai-chat-visual-b3f48351-6aab-49c9-bd9f-3d5cb86fdfd1/image1_iconagent300x300.png

AI chat visual

by Jesper forretning

Download SampleInstructions

An AI-powered chat interface for interacting with your Power BI report data through Implement's GenA

Conversational analytics inside Power BI The AI Chat Visual brings a secure conversational interface into Power BI reports, allowing authorised users to ask natural-language questions about report data and receive structured analytical responses from a configured GenAI backend service. The visual is designed as a thin client embedded directly in Power BI. It authenticates the signed-in Power BI user, sends the prompt and available report context to a configured backend endpoint, and displays the returned response inside the report. Intended use and access restrictions This visual is intended for use by Implement Consulting Group and parties explicitly authorised to use Implement's GenAI service. The visual itself is a thin Power BI client and does not include a public backend service. To use the chat functionality, the visual must be configured with a compatible backend endpoint that handles authentication, authorisation, prompt processing, and response generation. Implement's backend service is access-controlled and available only to authorised users, organisations, or tenants. Users outside the authorised organisation, tenant, or access group cannot use Implement's hosted GenAI backend. Organisations that want to use the visual independently would need to provide and configure their own compatible backend service. If an unauthorised user opens the visual or attempts to use a backend service they are not authorised for, the visual displays a restricted-access message explaining that the configured service is not available for the current user or tenant. No protected chat functionality or backend content is exposed to unauthorised users. How it works Authenticates the signed-in user using the Power BI Authentication API and Microsoft Entra ID. Sends the user's prompt, optional bound report context, and authentication token to the configured backend endpoint. The backend validates the user token and determines whether the user is authorised to access the configured service. Receives a response from the configured GenAI service when access is allowed. Displays the answer in the visual as text, tables, KPIs, charts, or query details, depending on the response. Configuration requirements The Power BI or Fabric tenant administrator must allow the Power BI Authentication API for the relevant scenario. The report author must configure the backend API URL in the visual's formatting pane. The configured backend endpoint must be one of the permitted hosts declared in the visual package. The signed-in user must be authorised to access the configured backend service. Optional report context fields can be bound to improve answer relevance, such as current user, report page, and active filters. Error handling and troubleshooting The visual displays specific error information directly in the UI to support troubleshooting. Error messages help distinguish between missing configuration, unavailable authentication, unauthorised access, backend connectivity issues, and service-side processing errors. Missing configuration: the backend API URL has not been configured. Authentication unavailable: the Power BI Authentication API is disabled, unsupported, or unavailable in the current environment. Unauthorised access: the signed-in user, organisation, or tenant is not authorised for the configured backend service. Backend unreachable: the configured backend endpoint could not be reached. Request failed: the backend returned a specific HTTP or service error. Security The visual is designed so that access is controlled through Microsoft Entra ID authentication and the packaged visual configuration. Permitted backend hosts and the Entra ID token audience are declared in the signed visual package. The backend validates incoming user tokens before processing requests. The visual should only be configured to connect to trusted backend services. Prompts, report context, and responses are processed according to the configured backend service and the user's authorisation. Support For access requests, configuration support, or interest in a similar conversational analytics solution, please use the support link associated with this offer.

Visual capabilities

When this visual is used, it
  • Can access external services or resources

At a glance

https://catalogartifact.azureedge.net/publicartifacts/jespers_forretning.ai-chat-visual-b3f48351-6aab-49c9-bd9f-3d5cb86fdfd1/image0_converted.png
English (South Africa)
Your Privacy Choices Opt-Out Icon Your Privacy Choices
Consumer Health Privacy Sitemap Contact Us Privacy & Cookies Terms of Use Trademarks About our ads Manage cookies