AI Readiness · 5
fabric.guru
Oct 2, 2026
4/5 · Tier A
AI ReadinessGovernanceAPI Automation
tutorial
<p><strong>Copilot access for semantic models</strong> can now be controlled separately for read-only users. Writers can enable or disable whether a model is discoverable and usable through Microsoft Copilot, report Copilot, and data agents.</p><p>The setting is enabled by default. The article demonstrates applying it from a Fabric notebook with SemPy and a REST client, but relies on an internal endpoint whose authentication and behavior may change. Restricting access prevents unprepared models from producing poorly grounded AI responses while leaving normal report consumption available.</p>
Programmatic Copilot gating improves AI governance, with concrete SemPy automation and an important unstable API warning.
Fabric Updates
Sep 29, 2026
4/5 · Tier A
AI ReadinessSemantic ModelGovernance
announcement
<p><strong>Fabric Planning</strong> connects budgets, forecasts, targets, and scenarios directly to existing governed semantic models. Planners can connect using their signed-in identity, retain existing permissions, write plan data back into the semantic foundation, and expose planning concepts to Fabric IQ's ontology and agents.</p><p>The preview <strong>Native Planning Engine</strong> uses Rust and Apache Arrow with OneLake integration, claiming up to 10x performance improvements and support for models with tens of millions of cells. Event triggers can start pipelines, Dataflow Gen2, notebook, refresh, Teams, or Outlook actions when writeback starts, succeeds, or fails, with writeback metadata passed as parameters.</p><p>The feature remains a planning workload with preview components and implied scale, governance, and writeback considerations. Organizations must validate calculation behavior, concurrency, permissions, trigger failure handling, and the operational cost of downstream automation before replacing a dedicated planning platform.</p>
Fabric Planning adds semantic-model writeback, ontology integration, scalable calculations, and event-driven automation with concrete preview limits.
Power BI Updates
Sep 29, 2026
3/5 · Tier B
AI ReadinessSemantic ModelReport Authoring
announcement
Power BI Desktop is gaining a natural-language app-building experience powered by Fabric apps. Users can start from a trusted semantic model, generate and edit an application, publish it, accept inputs, write back data, preserve shared state, and support operational workflows.
The preview will be available to Pro, Premium Per User, and Fabric capacity customers. Semantic views aim to define governed metrics closer to OneLake data, while Fabric IQ in Microsoft 365 Copilot Cowork and Chat is generally available. Fabric IQ MCP is also entering general availability for conversational analytics in development environments and external MCP clients.
The announcement provides direction rather than detailed limits, capacity costs, performance characteristics, or governance controls. Apache Ossie and cross-platform semantic conversion remain strategic interoperability efforts.
Introduces natural-language Fabric apps, semantic views, Copilot integrations, and MCP availability, but provides limited implementation detail.
databricks
Sep 28, 2026
2/5 · Tier B
AI Readiness
announcement
<p><strong>Lakebase Search</strong> adds full-text and vector search capabilities to Postgres for applications and AI agents. It targets retrieval workloads that traditional transactional databases handle poorly.</p><p>The topic is relevant to AI-ready architecture, but the extract provides no limits, pricing, indexing behavior, or comparison with Fabric search and retrieval options.</p>
Postgres vector and full-text search is adjacent AI infrastructure, but not directly a Fabric capability or migration analysis.
Tabular Editor
Sep 24, 2026
3/5 · Tier B
AI ReadinessSemantic ModelGovernance
deep-dive
<p><strong>Tabular Editor 3.27</strong> includes an MCP server that lets coding agents such as Claude Code, GitHub Copilot, Codex, and Cursor read and edit the model currently open in Tabular Editor, including unsaved changes.</p><p>Permissions separate metadata read/write, model-data access, Best Practice Analyzer execution, documents, and macros. Changes remain visible in the TOM Explorer and Properties pane, each script becomes an undo step, and scripts attempting operations outside the model require manual review. Agents can also use BPA rules, VertiPaq Analyzer statistics, DAX queries when permitted, and version-specific C# scripting references.</p>
Live MCP editing adds practical agent controls, permissions, auditability, and review workflows for semantic-model development.
Breaking Change · 18
fabricknownissues
Sep 29, 2026
5/5 · Tier A
⚠️
Breaking ChangeSpark RuntimeGovernance
release-notes
<p><strong>Spark Runtime 2.0</strong> can cause materialized lake view refreshes to fail across workspaces. Dependency views using <code>SELECT *</code> may return <code>Spark_System_OneLakeSecurity_OperationFailed</code>.</p><p>The issue is active and combines a runtime dependency with OneLake security-operation failures, so upgrading or standardizing runtime versions requires validation rather than blind rollout. Existing MLV refresh schedules may remain broken until Microsoft provides a fix or a documented workaround.</p>
Active Fabric known issue breaks MLV refreshes and OneLake security operations on Spark Runtime 2.0.
fabricknownissues
Sep 29, 2026
2/5 · Tier B
⚠️
Breaking ChangeAI Readiness
release-notes
<p>Generating an Operations Agent playbook from an Ontology in the new experience can time out. The issue is active and may prevent the playbook artifact from being created.</p><p>The extract gives no scope, timeout threshold, workaround, or cost impact, so the practical effect is limited to teams using Ontology-backed Operations Agent authoring.</p>
Active IQ known issue blocks Operations Agent playbook generation, but affects a narrower preview-style workflow.
fabricknownissues
Sep 29, 2026
3/5 · Tier B
⚠️
Breaking ChangeAI ReadinessSemantic Model
release-notes
<p>Fabric Data Agents cannot currently use an Ontology as context when that Ontology is backed by a semantic model. In mixed-source Ontologies, the agent uses only non-semantic-model sources.</p><p>This constrains grounding designs that expect business definitions, measures, or governed relationships from a Power BI semantic model to flow through an Ontology into a Data Agent.</p>
Active limitation prevents Data Agents from using semantic-model-backed Ontology context, affecting grounding architecture.
fabricknownissues
Sep 29, 2026
2/5 · Tier B
⚠️
Breaking ChangeReal-Time IntelligenceAI Readiness
release-notes
<p>Real-Time Dashboards may fail to create the expected visuals when an Ontology is attached. Simple one-entity Ontologies may work, while more complex Ontology scenarios can fail.</p><p>The issue affects dashboard authoring and visual generation, not necessarily the underlying real-time data or Ontology. The extract does not identify a workaround or supported complexity boundary.</p>
Active IQ issue can prevent Real-Time Dashboard visual creation from complex Ontologies.
fabricknownissues
Sep 29, 2026
3/5 · Tier B
⚠️
Breaking ChangeAPI AutomationGovernance
release-notes
<p><strong>Ontology MCP</strong> does not currently authenticate with service principals. Automated workloads therefore cannot use this MCP endpoint with an application identity.</p><p>Interactive user authentication may still work, but unattended pipelines, managed services, and cross-environment automation must use an alternative pattern or wait for support.</p>
Active MCP authentication limitation blocks service-principal automation and affects noninteractive security design.
fabricknownissues
Sep 29, 2026
3/5 · Tier B
⚠️
Breaking ChangeMigrationAI Readiness
release-notes
<p>Ontology migration to the new experience fails when entity types use <strong>composite keys</strong>. Existing models with multiple columns forming an entity identifier cannot complete the migration as configured.</p><p>The practical consequence is a migration dependency: teams must identify affected entity types and determine whether keys can be remodeled before scheduling cutover.</p>
Composite-key support failure can block Ontology migration and requires redesign or remediation before adopting the new experience.
fabricknownissues
Sep 29, 2026
3/5 · Tier B
⚠️
Breaking ChangeMigrationAI Readiness
release-notes
<p>Ontology migration to the new experience fails when properties are configured as <strong>Defined at Binding</strong>. Existing ontologies using that property configuration require remediation before migration can complete.</p><p>This creates a model-compatibility gate rather than a transient user-interface issue: migration planning must include an inventory of these properties and a tested replacement design.</p>
Defined-at-binding properties can block Ontology migration, creating a concrete compatibility gap for existing models.
fabricknownissues
Sep 29, 2026
3/5 · Tier B
⚠️
Breaking ChangeAI ReadinessAPI Automation
release-notes
<p><strong>Ontology Agent</strong> and Ontology MCP cannot query ontology rules through <code>ask_ontology</code>. Users can therefore receive answers based on accessible ontology data without the rules being available as part of the query path.</p><p>This limits rule-aware agent behavior and prevents integrations from relying on <code>ask_ontology</code> for complete ontology reasoning until rule access is supported.</p>
Ontology Agent and MCP cannot query ontology rules, limiting governed reasoning and automated rule-aware answers.
fabricknownissues
Sep 28, 2026
3/5 · Tier B
⚠️
Breaking ChangeAI ReadinessGovernance
release-notes
<p><strong>Microsoft Copilot Chat</strong> may silently ignore Power BI URLs and answer from general search instead of Power BI data. The active issue affects customers who have not enabled <strong>Share Fabric data with your Microsoft 365 services</strong>.</p><p>Cross-geography Fabric and Microsoft 365 tenants require administrator acknowledgement of data sharing. Microsoft states that enabling this setting is not intended to be a requirement, so the behavior represents a product issue rather than a documented prerequisite.</p>
Active Copilot grounding issue depends on a tenant setting and cross-geography data-sharing configuration.
fabricknownissues
Sep 28, 2026
4/5 · Tier A
⚠️
Breaking ChangeDelta & ParquetMigration
release-notes
<p>The September 2026 <strong>On-premises data gateway</strong> release, version 3000.334.4, may cause Fabric Copy activities and Copy Jobs to fail when performing <strong>UPSERT</strong> operations into existing Lakehouse tables.</p><p>Initial loads into empty tables can succeed because they avoid the subsequent Delta-table index-building path. Affected runs may return a missing or mismatched <strong>System.IO.Hashing</strong> assembly error, indicating a gateway dependency problem rather than bad source data.</p>
A version-specific gateway regression can break Lakehouse UPSERT operations after initial loads.
fabricknownissues
Sep 28, 2026
4/5 · Tier A
⚠️
Breaking ChangeCI/CDSemantic Model
release-notes
<p><strong>Power BI Deployment Pipelines</strong> may fail when promoting semantic models that require purge-data approval. The active error is <strong>Alm_InvalidRequest_PurgeRequired</strong>; existing datasets and reports remain accessible, but the deployment cannot complete.</p><p>The issue affects the promotion operation rather than the already-deployed content. Microsoft provides a workaround through the Learn documentation's <strong>Try It</strong> function or an authenticated CLI invocation.</p>
Deployment Pipelines can block semantic-model promotion when purge-data approval is required.
fabricknownissues
Sep 28, 2026
4/5 · Tier A
⚠️
Breaking ChangeDelta & ParquetWarehouse
release-notes
<p><strong>Fabric Data Warehouse</strong> tables using <strong>Data Clustering</strong> may fail when read by some external Delta Lake clients. Those readers may not understand the clustering metadata written by Warehouse and can return deserialization errors.</p><p>Queries through Fabric experiences such as Lakehouses may continue to work, but metadata operations including <strong>DESCRIBE DETAIL</strong> can still fail. The problem is therefore an interoperability boundary, not necessarily corruption of the table itself.</p>
Warehouse Data Clustering metadata can break interoperability with external Delta Lake readers.
fabricknownissues
Sep 26, 2026
3/5 · Tier B
⚠️
Breaking ChangeGovernanceWarehouse
release-notes
<p>Data Factory connection creation may fail when a workspace contains only <strong>Private Link</strong>-enabled Warehouses, even when a virtual network data gateway and private endpoint are configured. The failure occurs from Get data or Manage connections and gateways.</p><p>Test connection uses the Warehouse public fully qualified domain name instead of the workspace FQDN, producing a generic connectivity, timeout, or credentials error. This makes the issue appear to be authentication or firewall-related when endpoint selection is the underlying problem.</p>
Private Link-enabled Warehouse connections can ignore configured VNet routing and target an inaccessible public endpoint.
fabricknownissues
Sep 25, 2026
4/5 · Tier A
⚠️
Breaking ChangeWarehouseGovernance
release-notes
<p><strong>Copy and CopyJob writes to Data Warehouse fail</strong> when the target workspace is protected by Private Link in specific connectivity configurations.</p><p>With gateway-based movement, externally staged Azure Storage using Workspace Identity can fail when storage firewall rules do not allow the Warehouse IPs. Cloud-run sources can fail when the Warehouse connection uses Private Link.</p><p>Workarounds are to allowlist Warehouse IPs on the staging storage firewall, or run the source connection through the same gateway as the Warehouse connection. Engineering is investigating; this is not a general Copy limitation.</p>
Active private-link defect breaks Copy and CopyJob writes to Fabric Data Warehouse.
fabricknownissues
Sep 25, 2026
4/5 · Tier A
⚠️
Breaking ChangeWarehouseGovernance
release-notes
<p><strong>Pipeline and Copy Job connection creation can fail for public Warehouses</strong> when the same user can also access Warehouses in Private Link-enabled workspaces.</p><p>The reported error claims the data source is inaccessible, timed out, or has invalid credentials, even when the intended public Warehouse is otherwise reachable. The issue occurs during connection creation and is tied to the user's mixed access context.</p>
Active connection-discovery defect can block pipeline access to public Warehouses across mixed Private Link environments.
fabricknownissues
Sep 25, 2026
4/5 · Tier A
⚠️
Breaking ChangeWarehouseGovernance
release-notes
<p><strong>Private Link-enabled Warehouses may not appear selectable</strong> when configuring a pipeline or Copy Job. The Warehouse picker can show an empty error screen during the Get data or Copy Job experience.</p><p>This blocks configuration through the graphical selection flow; it does not establish that the Warehouse itself is unavailable. The issue remains active.</p>
Active picker defect prevents selecting Private Link-enabled Warehouses in pipelines and Copy Jobs.
fabricknownissues
Sep 25, 2026
3/5 · Tier B
⚠️
Breaking ChangeWarehouse
release-notes
<p><strong>Copy Job full loads can fail or produce unexpected results</strong> when <strong>Truncate after load</strong> is enabled and a precreated target table has a different column order from the source.</p><p>The risk applies even when source and target column names are logically compatible, because this scenario can cause the load behavior to depend on ordinal position. Aligning target column order with the source avoids the documented configuration.</p>
Active Copy Job full-load defect can misalign data when precreated target columns differ from source order.
fabricknownissues
Sep 23, 2026
2/5 · Tier B
⚠️
Breaking ChangeReport Authoring
release-notes
<p><strong>Existing conditional formatting for a Card visual's Callout Label may disappear from the format pane</strong> after opening the report in the Power BI service or newer Desktop versions.</p><p>The underlying <strong>Fx</strong> rule continues to work, and report rendering is not affected. The defect primarily limits inspection or editing of formatting configured in July 2026 or earlier.</p>
Active Power BI Card formatting-pane defect hides existing callout-label rules while rendering remains correct.
Semantic Model · 6
<p><strong>Fabric-RLM</strong> connects a Fabric notebook to one or more Power BI semantic models and lets an LLM inspect schemas, measures, relationships, and period completeness before querying data. Its semantic-model skill can aggregate measures, execute DAX, correct generated-code errors, and record which model fields supported each result.</p><p>The workflow extends beyond conversational answers: a task can produce HTML or Markdown reports and Excel workbooks. Runs can be inspected step by step, including model responses, generated code, outputs, and metrics.</p><p>Accuracy depends on explicitly designed validation. Independent queries can recompute headline and total revenue, reject incomplete comparison periods, verify category names, enforce report structure, and return failed work to the model for repair. The example requires a paid Fabric capacity and uses GPT-5.1 with high reasoning effort, so execution and model costs require platform controls.</p>
Practical notebook automation combines semantic-model querying, LLM reasoning, artifact generation, lineage, and independent validation.
<p><strong>Apache Ossie</strong> defines a vendor-neutral format for exchanging semantic metadata across analytics, BI, and AI platforms. Microsoft and Snowflake are collaborating on converting one Ossie document into both a <strong>Snowflake Semantic View</strong> and a <strong>Power BI semantic model</strong>.</p><p>The integration is intended to preserve shared business meaning without moving or duplicating underlying data. Microsoft also plans to contribute semantic-model expertise, including potential recognition of <strong>DAX</strong> and support for ontologies. Ossie remains an Apache incubating project, so its specification, governance, tooling, and compatibility are still evolving.</p>
Apache Ossie enables portable semantic definitions across Power BI and Snowflake, but provides limited implementation detail.
<p>40 features - 12 must-know - 15 worth knowing.</p><ul><li>Deprecation of Old File Picker experience in Power BI Desktop</li><li>Power BI Projects (PBIP) (Generally Available)</li><li>New Purge Data option for Fabric APIs</li><li>Power BI Agentic Experiences (Generally Available)</li><li>Power BI data answers in Microsoft Copilot Chat (Generally Available)</li><li>Fabric IQ Plugin in Microsoft 365 Copilot Cowork (Generally Available)</li><li>New control for Copilot access to semantic models</li><li>Remote Power BI Authoring MCP Server (Preview)</li><li>Create and edit DirectQuery semantic models in the Power BI service</li><li>Approximate Distinct Count comes to Import and Direct Lake (Preview)</li><li>Accelerate text searches in your semantic models with String Indexing (Preview)</li><li>Introducing Full-Text Search in Power BI Semantic Models (Preview)</li></ul>
12 must-know of 40 features
<p>9 features - 2 must-know - 7 worth knowing.</p><ul><li>Fabric data connectors</li><li>Built to be governed</li></ul>
2 must-know of 9 features
<p><strong>Fabric IQ</strong> is generally available in Microsoft 365 Copilot Cowork and Copilot Chat, grounding answers in existing governed Power BI reports and semantic models rather than a separate AI data copy. Users can reuse established measures, relationships, and certified definitions, while enhanced controls govern which models Copilot may access.</p><p>The release also previews GPU-accelerated Fabric Data Warehouse query execution and introduces early semantic views for defining reusable business meaning directly alongside OneLake data. Fabric IQ MCP is generally available, allowing development environments and external clients to access governed semantic content, while Fabric Apps extend trusted models into applications that can accept input, write back data, and preserve shared state.</p>
Combines Fabric IQ GA, GPU warehouse preview, semantic views, MCP governance, and agentic application capabilities.
<p><strong>Tabular Editor 3.27</strong> integrates agentic semantic-model development through a built-in MCP server and a generally available AI Assistant. Both can apply C#-scripted changes, but permissions, safety checks, visibility, selective revert, and undo controls keep edits reviewable. AI Assistant direct changes are opt-in and can use locally hosted models.</p><p>The release also adds unsaved change indicators, property-level revert, deleted-object restoration, improved Data Preview, user-context-aware calculated-column validation, and compatibility level 1706 support. A breaking change affects scripts: query and command calls on <code>Model.Database</code> must use built-in script helpers instead.</p>
The release combines governed AI model editing, granular change management, validation improvements, and a script-breaking API change.
