Microsoft intelligence layer
Work IQ, Fabric IQ & Foundry IQ
Microsoft introduced three intelligence layers at Ignite 2025 to give copilots, agents, and analytics better context. They are complementary, not interchangeable, and most organizations need help deciding what each one does and where it fits. Understanding these layers is the fourth stage of the AI Advantage Framework.
What these layers are
Work IQ, Fabric IQ, and Foundry IQ are architectural labels for three types of enterprise intelligence. Each one provides a different kind of context to AI agents and copilots.
Work IQ
Intelligence about how work happens
Work IQ captures collaboration patterns, communication signals, meeting behavior, document usage, and personal preferences from Microsoft 365. It powers Copilot's ability to personalize responses and understand organizational context. Think of it as the intelligence layer built from emails, chats, meetings, and files your organization already creates every day.
Fabric IQ
Intelligence from enterprise data
Fabric IQ unifies structured data across OneLake and organizes it with business semantics through ontologies and semantic models. It extends Power BI's trusted definitions into operations and AI, giving agents the ability to understand business metrics, not just raw numbers. Fabric IQ includes an ontology engine, semantic model integration, a graph engine, data agents, and operations agents.
Foundry IQ
Intelligence from knowledge and reasoning
Foundry IQ is a managed knowledge retrieval system for grounding AI agents. It federates data across Microsoft 365, SharePoint, Fabric IQ, Azure Blob Storage, Azure AI Search indexes, and the web. It handles permission-aware retrieval, multi-source reasoning, and enterprise governance through Entra ID and Microsoft Purview integration.
How they differ
The three layers serve different purposes and draw from different sources. This table makes the boundaries concrete.
| Work IQ | Fabric IQ | Foundry IQ | |
|---|---|---|---|
| Primary purpose | Work context and collaboration intelligence | Business data semantics and analytics intelligence | Enterprise knowledge grounding and retrieval |
| Best-fit context | Emails, meetings, chats, documents, organizational signals | Structured data, semantic models, ontologies, business metrics | Documents, policies, contracts, knowledge bases, web content |
| Best-fit use case | Copilot personalization, workflow pattern detection, collaboration insights | Consistent analytics, data agents, operations agents, unified business definitions | AI agent grounding, knowledge retrieval, multi-source reasoning |
| Common confusion | Confused with Copilot itself; Work IQ powers Copilot, it is not Copilot | Confused with standard Power BI; Fabric IQ extends semantic models into operations and AI | Confused with basic Azure AI Search; Foundry IQ adds governance, permissions, and federated retrieval |
| Related Microsoft surface | Microsoft 365, Copilot, Outlook, Teams, SharePoint | Microsoft Fabric, Power BI, OneLake | Microsoft Foundry (formerly Azure AI Foundry), Azure AI Search |
How they connect to Copilot, Fabric, and Foundry
These layers are not standalone products. They are intelligence capabilities that enhance existing Microsoft platforms.
Work IQ powers Copilot
Work IQ is the intelligence engine behind Microsoft 365 Copilot's personalization, memory, and organizational awareness. Copilot without Work IQ is a general assistant. Copilot with Work IQ understands your work patterns, your organization's structure, and your preferences.
Fabric IQ extends Fabric
Fabric IQ sits inside Microsoft Fabric as a workload. It takes the data already unified in OneLake and adds semantic meaning through ontologies and enhanced semantic models. It turns Fabric from a data platform into an intelligence platform where agents understand business context.
Foundry IQ grounds Foundry
Foundry IQ is a preview capability of Microsoft Foundry (formerly Azure AI Foundry). It provides the knowledge retrieval and grounding layer that AI solutions built in Foundry need to answer questions accurately, respect permissions, and cite sources.
Together, they form the intelligence layer
When combined, these three layers give AI agents access to work context (Work IQ), business data semantics (Fabric IQ), and enterprise knowledge (Foundry IQ). This is what Microsoft means by a unified intelligence layer for the enterprise.
What buyers usually get wrong
Assuming one layer replaces the others
Each layer serves a different purpose. Work IQ does not replace the need for Fabric IQ's data semantics, and Foundry IQ does not eliminate the need for Work IQ's collaboration context. Organizations that treat them as interchangeable end up with gaps.
Thinking product names define architecture
Work IQ, Fabric IQ, and Foundry IQ are architectural concepts, not standalone SKUs you purchase separately. Understanding what each layer does matters more than tracking product names, which Microsoft updates frequently.
Buying before a grounding strategy exists
The most common mistake is investing in AI agents or copilot deployments without first deciding which intelligence layers need to be in place. Without that grounding strategy, agents produce generic or inaccurate results.
Choose your path
Explore the specific question that matters most for your team right now.
Part of a larger system
Why these questions matter beyond Microsoft
Understanding Microsoft's intelligence layers is the fourth stage of the AI Advantage Framework. The first three stages address which work to fund, whether the information is usable, and whether the workflows can execute. Platform decisions make more sense when those foundations are in place.
AI Advantage Framework
AI Fit & Governance → AI-Ready Data → Operational AI → Microsoft Intelligence
Need help deciding where these layers fit?
Microsoft's intelligence layers are evolving fast. Marquee Insights helps organizations separate what is ready now from what requires patience, and decide where to invest first.