Skip to content

FAQ⚓︎

Frequently asked questions about Portal AI.

General⚓︎

What is Portal AI?⚓︎

Portal AI is an AI-powered assistant built into PowerBI Portal that lets users ask natural language questions about the data in their Power BI reports. It connects to the semantic model (previously referred to as a dataset) behind each report, generates DAX queries, and returns conversational answers based on real data.

Which AI model does Portal AI use?⚓︎

Portal AI uses AI models hosted in Azure AI Foundry to understand questions and generate answers from report data.

Does Portal AI have access to my raw data?⚓︎

Portal AI queries your data through the Power BI semantic model using DAX, the same query language Power BI uses internally. It does not access your underlying data sources (SQL databases, files, etc.) directly. All queries go through Power BI's standard security and access layer.

In which languages can I ask questions?⚓︎

Portal AI supports questions in any language. However, the quality of responses may vary depending on the language. Results tend to be best in English and other widely spoken languages.

Usage Credits⚓︎

Do Portal AI interactions consume credits?⚓︎

Yes. Portal AI interactions consume credits depending on the query and response.

What happens when I run out of credits?⚓︎

When credits are exhausted, users will see a "Usage limit reached" message in the chat window. Portal AI will no longer respond to questions until credits are available again.

Where can I review usage?⚓︎

Usage is recorded in Usage History in License & Billing.

Row-Level Security (RLS)⚓︎

Does Portal AI work with reports that use RLS?⚓︎

Yes. When a report has RLS configured, Portal AI respects the RLS rules defined for that report in PowerBI Portal. Users only receive answers based on the same data they can see when viewing the report normally.

What are the requirements for RLS to work with Portal AI?⚓︎

Reports with RLS require a Premium or Embedded capacity. Without capacity, Portal AI will display an error message for RLS-enabled reports.

Schema Discovery⚓︎

What is schema discovery?⚓︎

Schema discovery is the process of reading the structure (tables, columns, measures, and relationships) from your Power BI semantic model. Portal AI needs this schema to understand your data and generate accurate DAX queries.

When should I refresh the schema?⚓︎

Refresh the schema whenever significant changes are made to your semantic model, such as adding or removing tables, columns, or measures, changing relationships, or renaming fields. If the schema is outdated, Portal AI may not be aware of new fields or may reference columns that no longer exist.

Schema discovery failed. What should I do?⚓︎

Common causes of schema discovery failures:

  1. Power BI permissions not configured: Ensure all required settings are enabled in the admin portal in Power BI (see Enabling Portal AI).
  2. Permissions not propagated yet: After changing settings in the admin portal in Power BI, wait up to 15 minutes and try again.
  3. Service principal access: The PowerBI Portal service principal must have access to the workspace containing the report.

What does the semantic model too large error mean?⚓︎

Portal AI validates the semantic model schema size before it can be used. If this error appears, the model contains more metadata than Portal AI can process. Consider hiding unnecessary tables, columns, and measures in Power BI Desktop, then refresh the schema again.

Troubleshooting⚓︎

Portal AI answers are not accurate⚓︎

  • Make sure the schema is up to date. Refresh it if you've recently changed the semantic model.
  • Be specific in your questions. Reference the actual column and measure names from your report.
  • Clear the conversation and start fresh if the context has become confusing.
  • Remember that Portal AI uses AI and may occasionally make mistakes. Always validate important data.

The chat button does not appear on a report⚓︎

This can happen for several reasons:

  • Portal AI is not enabled for the environment. Go to the Portal AI tab in Settings.
  • Portal AI is not enabled for the specific report. Edit the report and open the Portal AI tab.
  • The administrator has not granted you AI access for that report.

Portal AI says "capacity unavailable"⚓︎

Portal AI needs to connect to Power BI to execute queries. If the Power BI capacity is paused or not responding, Portal AI will show this error. The system will attempt to resume the capacity automatically, but if it times out, try again after a few minutes.