Microsoft brings Power BI semantic models, metrics and rules to Copilot Chat and Cowork. Understand the gain and what still needs validation.

Direct answer

Microsoft announced on September 28, 2026 the general availability of Fabric IQ in Copilot Chat and Cowork. The feature uses semantic models, metrics, relationships, and business definitions governed in Power BI to inform responses and work steps. This reduces disputes over which number is official and preserves existing access controls, but it doesn't automatically fix bad data, ambiguous metrics, or outdated semantic models. The quality of the assistant remains dependent on the quality and governance of the data layer.

Semantics transforms data into context

Relationships, metric names and calculation rules help the assistant respond according to the company's language, not just due to textual similarity.

A governed source reduces competing numbers

When the report and wizard use the same template, the discussion can move from the number's origin to its interpretation and action.

Access control remains essential

Grounding does not legitimately extend permissions. Sensitive groups, roles, and data must continue to be reviewed in Power BI and Fabric.

Semantic model can also make mistakes

Old settings, implicit filters, or insufficient source quality will play faster. Validation and ownership are mandatory.

Nexus Reading

Start with a few critical questions, associate each answer with metric, owner, and source, and test for consistency before expanding Copilot access.

FAQ

Is Fabric IQ available now?

Microsoft reports general availability for Copilot Chat and Cowork on September 28, 2026.

What data supports the answers?

Power BI semantic models, metrics, relationships, and governed business definitions.

Does the resource guarantee correct answers?

No. It improves consistency and traceability, but it depends on the quality of the data and the semantic model.

Essential guides to delve deeper into the decision

Primary sources and references

This editorial analysis was produced by Nexus from the official sources below, consulted on October 1, 2026. The text is original and interprets practical implications for companies.

Date reported by the main source: September 28, 2026.