Microsoft Foundry IQ is a knowledge layer for AI agents that creates reusable, multi-source knowledge bases, uses language-model-driven agentic retrieval to issue parallel subqueries and aggregate results, and returns permission-trimmed, traceable, grounded outputs.
Published May 16, 2026
Foundry IQ is labeled as preview; it is provided without an SLA. Capabilities and behavior may change as the product evolves.
Query planning for agentic retrieval is described as automated and not customizable in the managed offering.
Agentic retrieval and related premium features are available in selected regions and may depend on higher-tier capabilities such as semantic ranker.
Practitioners have reported confusion about portals and SDKs after product renames and packaging changes; this increases adoption cost but is not a functional defect per se.
Foundry IQ is described in documentation as a public preview. Preview services typically have fewer mature third-party reviews and limited product listings specific to Foundry IQ. To broaden coverage, this section references reviews of Azure AI Search (the underlying platform), community discussions, and Microsoft-authored evaluations.
G2 (Azure AI Search reviews):
Gartner Peer Insights: a listing exists for Azure AI Search reviews and ratings; access to full content may be gated.
Community discussions often focus on product renaming, SDK choices, and ecosystem complexity rather than long-term production experience. Observed community themes include uncertainty about which SDKs to use after renaming or repackaging and questions about what is new versus rebranded Azure AI Search features.
Organizations standardized on Microsoft services such as SharePoint, OneLake, Fabric, and Entra identity can reduce the need for custom security integration by using ACL-aware retrieval and Purview label enforcement. The ability to offer a single knowledge base endpoint that multiple agents can reuse is also a common rationale.
A vendor-authored analysis reported improved end-to-end RAG answer scores when using agentic retrieval versus a brute-force approach across sources. Organizations should request dataset characteristics, scoring rubrics, and reproducibility details before relying on vendor-reported improvements.
Most suitable for mid-market to enterprise organizations with complex content and access control needs, particularly where Microsoft 365 and Azure are already in use.
Regulated and knowledge-intensive industries: finance, healthcare, manufacturing, professional services, and public sector—due to document-level permissioning and governance requirements.
Get a comprehensive analysis of Microsoft Foundry IQ including market position, competitive landscape, adoption trends, and peer benchmarks.
Download full report →Microsoft Foundry IQ is described as a knowledge layer for AI agents built on Azure AI Search. It focuses on creating reusable, multi-source knowledge bases that can be connected to multiple agents so agents can retrieve permission-trimmed enterprise knowledge and return grounded outputs with traceability.
A key differentiator emphasized by Microsoft is agentic retrieval: rather than a single query to a single index, Foundry IQ uses a language model–driven planning step to generate subqueries, run parallel searches across sources, and aggregate results.