Forum Discussion
Expression.Error: The import AIFunctions.PostProcess matches no exports. Did you miss a module refer
- 9 months ago
Hi agoldberg
This is AI generated and unverified, but I hope it helps:
Your issue is happening because Microsoft retired Cognitive Services and Azure Machine Learning integration in Power BI as part of the transition to Microsoft Fabric AI services. This retirement was completed on September 15, 2025, and all related features (including AI Insights like Extract Key Phrases and Score Sentiment) were removed from Power BI Desktop and Service.
Here’s what changed and why your M code is failing:
✅What Happened
- AIFunctions.PostProcess and AIFunctions.Contents were part of the old AI Insights connector that relied on Cognitive Services.
- Microsoft announced that starting August 11, 2025, you could no longer create new models using these services, and by September 15, 2025, refresh support for existing models ended. After that date, the functions were completely removed from Power BI. [m365admin....sontek.net], [powerbi.mi...rosoft.com], [datachant.com]
✅Why Your Reports Broke
- Any query steps using AIFunctions.PostProcess or AIFunctions.Contents now throw Expression.Error because those functions no longer exist in the Power Query engine.
- This affects both Premium capacity and Pro users—the retirement was global.
🔄What You Can Do
Microsoft recommends migrating to Fabric AI services or Azure AI REST APIs. Here are your options:
Move to Microsoft Fabric AutoML or AI Services
- Use Synapse Data Science in Fabric for AutoML workflows.
- For text analytics (sentiment, key phrases), use Azure AI Language integrated with Fabric. [datachant.com]
Call Azure AI REST APIs from Power Query
- Replace the old M code with custom calls to Azure Cognitive Services endpoints.
- Example approach:
- Use Web.Contents() in M to call the Azure Text Analytics API.
- Parse the JSON response and extract sentiment or key phrases.
- This keeps the logic inside Power Query without moving to Fabric immediately.
Short-Term Workaround
- If migration to Fabric is not possible yet, you can implement sentiment analysis and key phrase extraction using:
- Azure AI Language REST API
- Or Python/R scripts in Power BI (requires enabling scripting).
- If migration to Fabric is not possible yet, you can implement sentiment analysis and key phrase extraction using:
--------------------------------
I hope this helps, please give kudos and mark as solved if it does!
Connect with me on LinkedIn.
Subscribe to my YouTube channel for Fabric/Power Platform related content!
Hi agoldberg
This is AI generated and unverified, but I hope it helps:
Your issue is happening because Microsoft retired Cognitive Services and Azure Machine Learning integration in Power BI as part of the transition to Microsoft Fabric AI services. This retirement was completed on September 15, 2025, and all related features (including AI Insights like Extract Key Phrases and Score Sentiment) were removed from Power BI Desktop and Service.
Here’s what changed and why your M code is failing:
✅What Happened
- AIFunctions.PostProcess and AIFunctions.Contents were part of the old AI Insights connector that relied on Cognitive Services.
- Microsoft announced that starting August 11, 2025, you could no longer create new models using these services, and by September 15, 2025, refresh support for existing models ended. After that date, the functions were completely removed from Power BI. [m365admin....sontek.net], [powerbi.mi...rosoft.com], [datachant.com]
✅Why Your Reports Broke
- Any query steps using AIFunctions.PostProcess or AIFunctions.Contents now throw Expression.Error because those functions no longer exist in the Power Query engine.
- This affects both Premium capacity and Pro users—the retirement was global.
🔄What You Can Do
Microsoft recommends migrating to Fabric AI services or Azure AI REST APIs. Here are your options:
Move to Microsoft Fabric AutoML or AI Services
- Use Synapse Data Science in Fabric for AutoML workflows.
- For text analytics (sentiment, key phrases), use Azure AI Language integrated with Fabric. [datachant.com]
Call Azure AI REST APIs from Power Query
- Replace the old M code with custom calls to Azure Cognitive Services endpoints.
- Example approach:
- Use Web.Contents() in M to call the Azure Text Analytics API.
- Parse the JSON response and extract sentiment or key phrases.
- This keeps the logic inside Power Query without moving to Fabric immediately.
Short-Term Workaround
- If migration to Fabric is not possible yet, you can implement sentiment analysis and key phrase extraction using:
- Azure AI Language REST API
- Or Python/R scripts in Power BI (requires enabling scripting).
- If migration to Fabric is not possible yet, you can implement sentiment analysis and key phrase extraction using:
--------------------------------
I hope this helps, please give kudos and mark as solved if it does!
Connect with me on LinkedIn.
Subscribe to my YouTube channel for Fabric/Power Platform related content!