Error message when trying to prep model for AI

This error looks service-side. “NullSchemaPackageEvenAfterForcedUpdate” with HTTP 500 usually means Power BI failed to generate or retrieve the AI schema package for the semantic model, even after forcing a refresh.

Try these in order.

1. Confirm the basics are truly met

• The semantic model sits in a Copilot-enabled workspace on supported capacity.

• You have authoring rights on the semantic model (Member/Contributor in the workspace and Build permission on the model). The Prep data for AI experience requires authoring permission, not Viewer access.

• The model refresh completes successfully (no refresh failures, no credential prompts).

2. Force a clean “schema + metadata” update from Desktop, then republish
   This often bypasses a broken service-side package.

• Update Power BI Desktop to the latest.

• Open the PBIX that publishes this semantic model.

• Refresh, then publish again to the same workspace.

• After publish, open the semantic model in the service and try Prep data for AI again.

3. Try authoring Prep data for AI in Desktop instead of the service
   Microsoft supports authoring these features in both Desktop and Service. Desktop authoring sometimes succeeds when the service dialog fails.

• In Desktop: Home ribbon, Prep data for AI.

• Make a tiny change (add one AI instruction or one schema entry), save, publish.

• Then validate in the service.

4. Eliminate model metadata edge cases that trip schema packaging
   I would check for:

• Duplicate display names across tables/columns/measures (same visible name in different places).

• Hidden key columns used in relationships without a readable description.

• Columns with mixed data types or “Any” type coming from Power Query.

• Measure names or table names with unusual characters.
  Fixing duplicates and cleaning names is low effort and improves Copilot quality anyway.

5. Quick isolation test
   Create a minimal copy to see if the issue is model-specific or tenant/service-wide.

• Save a copy of the PBIX.

• Remove all but 1–2 tables and a couple measures.

• Publish to a new test workspace (same capacity).

• Run Prep data for AI.
  If the small model works, the issue sits in the original model metadata. If it fails everywhere, the issue is tenant/service-side.

6. Collect the right evidence and escalate with Microsoft support
   If none of the above works, you need the activity details from the failure and a diagnostic capture.

• When the error pops, copy the Activity Id / Correlation Id (if shown).

• Use the Copilot pane diagnostics option in Desktop (Download diagnostics) and attach it to the support request.

Thank you!
Proud to be a Super User!
📩 Need more help?
✔️ Don’t forget to Accept as Solution if this guidance worked for you.
💛 Your Like motivates me to keep helping

Thankyou, MJParikh for your response.

Hi TerYeah,

We appreciate your inquiry through the Microsoft Fabric Community Forum.

We would like to inquire whether have you got the chance to check the solution provided by MJParikh to resolve the issue. We hope the information provided helps to clear the query. Should you have any further queries, kindly feel free to contact the Microsoft Fabric community.

Thank you.

Hi TerYeah,

We are following up to see if what we shared solved your issue. If you need more support, please reach out to the Microsoft Fabric community.

Thank you.

Hi TerYeah,

Thank you for the update.

Based on our understanding, the error “NullSchemaPackageEvenAfterForcedUpdate” might occur when Power BI fails to generate the Copilot AI schema package for a specific semantic model during the “Simplify data schema” process. The issue might be caused by the size or metadata complexity of the model rather than an environmental problem. Additionally, the Prep data for AI button being disabled in Desktop may be due to your report using a Live Connection. In this architecture, the semantic model metadata can only be edited in the Service, not in Desktop.

Also, Copilot might rely on a simplified semantic model schema and recommends reducing schema complexity to improve grounding and packaging reliability. As a workaround, open the semantic model directly in the Power BI Service, go to Prep data for AI and select Simplify data schema, manually deselect most tables and fields, keeping only 1 to 2 primary fact tables and a small set of key measures. Apply the changes. This controlled reduction might allow the AI schema package to generate successfully.

If the issue persists even after schema simplification, we kindly request you to raise a support ticket using the provided link:Microsoft Fabric Support and Status | Microsoft Fabric so that the backend team can analyse the issue.

Additionally, please refer to the links below:
Prepare Your Data for AI - Power BI | Microsoft Learn
Enable Fabric Copilot for Power BI - Power BI | Microsoft Learn
Copilot in Power BI tutorial - Prepare semantic model for AI - Power BI | Microsoft Learn
Best practices to optimize Q&A - Power BI | Microsoft Learn
Overview of Copilot for Power BI - Power BI | Microsoft Learn

We hope the information provided helps to resolve the issue. Should you have any further queries, kindly feel free to contact the Microsoft Fabric community.

Thank you.

Hi TerYeah,

Thankyou for the update.

Based on our understanding, since the “Simplify data schema” dialog never loads and the process spins for 90 seconds before returning NullSchemaPackageEvenAfterForcedUpdate, this indicates that the failure occurs during the initial backend Copilot schema package generation step, before the simplification UI is rendered.

At this stage, we kindly request you to raise a support ticket using the link provided:Microsoft Fabric Support and Status | Microsoft Fabric so that the backend team can analyse the Copilot schema service logs, enabling engineering to investigate the packaging failure directly.

Should you have any further queries, please feel free to contact the Microsoft Fabric community.

Thank you.

Hi TerYeah,

We are following up to inquire whether you have raised the support ticket. If you have already done so, we kindly request you to share feedback regarding the issue raised. In case a solution has been provided, we would be grateful if you could share it with the community. This will assist others facing similar challenges and benefit the wider community.

If you require any further assistance, please feel free to connect with the Microsoft Fabric community.

Thank you.