--- title: Microsoft IQ Overview | daily.dev description: Microsoft IQ is an enterprise intelligence framework that provides AI agents with the right context, knowledge, and tools to perform tasks effectively. It... canonical: https://daily.dev/posts/microsoft-iq-overview-ivkokuknu twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Microsoft IQ Overview | daily.dev og:description: Microsoft IQ is an enterprise intelligence framework that provides AI agents with the right context, knowledge, and tools to perform tasks effectively. It... og:url: https://daily.dev/posts/microsoft-iq-overview-ivkokuknu og:image: https://api.daily.dev/og/posts/IVKoKuKNu.png og:image:alt: Microsoft IQ Overview og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Microsoft IQ Overview **[John Savill's Technical Training](https://daily.dev/sources/john-savill)** · 25 min read · 2 upvotes · 1 comments ## Summary Microsoft IQ is an enterprise intelligence framework that provides AI agents with the right context, knowledge, and tools to perform tasks effectively. It consists of four components: Work IQ (M365 collaboration data, relationships, and personalization), Fabric IQ (data virtualization layer unifying structured and unstructured enterprise data via semantic ontologies), Foundry IQ (curated AI-powered knowledge bases with prescriptive search), and Web IQ (low-latency Bing-powered internet search returning only relevant passages). The framework emphasizes giving each agent its own identity for least-privilege access, RBAC, data governance, and threat protection against prompt injection and jailbreaking. Agent 365 handles identity, data protection, and agent registry. The overall goal is to supply agents with precisely the relevant data needed — optimizing token usage and response quality — while maintaining enterprise-grade security and observability for continuous improvement. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What is the difference between Work IQ, Fabric IQ, Foundry IQ, and Web IQ in Microsoft's agent architecture? Work IQ focuses on M365 productivity data, personalization, and memory of collaboration patterns across emails, meetings, and documents. Fabric IQ virtualizes siloed enterprise data behind semantic ontologies without requiring data migration. Foundry IQ creates prescriptive, curated knowledge bases restricting which sources an agent can query. Web IQ uses Bing's index to return only the relevant passage from a search result, not the full page, minimizing latency and token usage. _Anyone comparing enterprise AI agent architectures can track how these frameworks evolve through daily.dev._ ### Why do AI agent architectures use multiple models instead of one large frontier model? Using multiple models lets an architecture match model size and cost to task requirements: a large frontier model handles deep reasoning, huge context, and heavy tool use, while a smaller, cheaper model handles low-latency, simple interactions. Since parameter count drives hardware cost and there is no single best model as models change frequently, staying model-flexible avoids being tied to one that could become unavailable or outdated. _Developers architecting multi-model AI agents can follow this kind of design reasoning on daily.dev._ ### Why is it recommended to send only relevant data to a large language model instead of using its full context window? Sending only the most relevant, highest-quality data rather than filling a million-token context window reduces both computational cost and token spend while improving answer quality, since models are now billed per token and relying on the model to sift through excessive context produces lower-quality outcomes than curating precisely what it needs beforehand. _Teams optimizing token usage and answer quality in LLM pipelines can keep up with these practices via daily.dev._ ## Community discussion Top comments from developers on daily.dev. **@agustinbarrientos** · 0 upvotes > I want every autonomous agent to have its own identity and least-privilege role before it touches company data. ## Similar posts on daily.dev - [Work IQ: The Foundational Intelligence Layer for Modern Enterprise AI](https://daily.dev/posts/work-iq-the-foundational-intelligence-layer-for-modern-enterprise-ai-klpdox7df) · C\# Corner · 1 upvotes · 0 comments --- Tags: [#ai](https://daily.dev/tags/ai), [#llm](https://daily.dev/tags/llm), [#microsoft](https://daily.dev/tags/microsoft), [#ai-agents](https://daily.dev/tags/ai-agents), [#rag](https://daily.dev/tags/rag) [View this post on daily.dev](https://daily.dev/posts/microsoft-iq-overview-ivkokuknu) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Microsoft IQ Overview","url":"https://daily.dev/posts/microsoft-iq-overview-ivkokuknu","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/microsoft-iq-overview-ivkokuknu"},"datePublished":"2026-07-20T14:10:49.013Z","dateModified":"2026-09-14T06:52:08.546Z","description":"Microsoft IQ is an enterprise intelligence framework that provides AI agents with the right context, knowledge, and tools to perform tasks effectively. It...","image":"https://i.ytimg.com/vi/LrIS9lcSbko/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/LrIS9lcSbko/sddefault.jpg","isAccessibleForFree":true,"articleSection":"John Savill's Technical Training","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"John Savill's Technical Training","logo":"https://media.daily.dev/image/upload/s--4rApg0fT--/f_auto,q_auto/v1780213220/logos/john-savill?_a=BAMAMiWQ0","url":"https://daily.dev/sources/john-savill"},"commentCount":1,"discussionUrl":"https://daily.dev/posts/microsoft-iq-overview-ivkokuknu","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":2},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":1}],"keywords":"ai,llm,microsoft,ai-agents,rag","timeRequired":"PT25M","video":{"@type":"VideoObject","name":"Microsoft IQ Overview","description":"Microsoft IQ is an enterprise intelligence framework that provides AI agents with the right context, knowledge, and tools to perform tasks effectively. It...","thumbnailUrl":"https://i.ytimg.com/vi/LrIS9lcSbko/sddefault.jpg","uploadDate":"2026-07-20T14:10:49.013Z","duration":"PT25M","url":"https://api.daily.dev/r/IVKoKuKNu","embedUrl":"https://www.youtube.com/embed/LrIS9lcSbko"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"John Savill's Technical Training","item":"https://daily.dev/sources/john-savill"},{"@type":"ListItem","position":3,"name":"Microsoft IQ Overview"}]} {"@context":"https://schema.org","@type":"WebPage","@id":"https://daily.dev/posts/microsoft-iq-overview-ivkokuknu","comment":[{"@type":"Comment","text":"I want every autonomous agent to have its own identity and least-privilege role before it touches company data.","datePublished":"2026-08-02T23:14:13.282Z","url":"https://daily.dev/posts/IVKoKuKNu#c-lWQ4bTE12","author":{"@type":"Person","name":"Agustin Barrientos","url":"https://daily.dev/agustinbarrientos","image":"https://media.daily.dev/image/upload/s--5ayxQnqn--/f_auto/v1788281802/avatars/avatar_wQYYVe5Tbj0NJ7C7qPoa8?_a=BAMAMicg0"}}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/microsoft-iq-overview-ivkokuknu#faq","mainEntity":[{"@type":"Question","name":"What is the difference between Work IQ, Fabric IQ, Foundry IQ, and Web IQ in Microsoft's agent architecture?","acceptedAnswer":{"@type":"Answer","text":"Work IQ focuses on M365 productivity data, personalization, and memory of collaboration patterns across emails, meetings, and documents. Fabric IQ virtualizes siloed enterprise data behind semantic ontologies without requiring data migration. Foundry IQ creates prescriptive, curated knowledge bases restricting which sources an agent can query. Web IQ uses Bing's index to return only the relevant passage from a search result, not the full page, minimizing latency and token usage. Anyone comparing enterprise AI agent architectures can track how these frameworks evolve through daily.dev."}},{"@type":"Question","name":"Why do AI agent architectures use multiple models instead of one large frontier model?","acceptedAnswer":{"@type":"Answer","text":"Using multiple models lets an architecture match model size and cost to task requirements: a large frontier model handles deep reasoning, huge context, and heavy tool use, while a smaller, cheaper model handles low-latency, simple interactions. Since parameter count drives hardware cost and there is no single best model as models change frequently, staying model-flexible avoids being tied to one that could become unavailable or outdated. Developers architecting multi-model AI agents can follow this kind of design reasoning on daily.dev."}},{"@type":"Question","name":"Why is it recommended to send only relevant data to a large language model instead of using its full context window?","acceptedAnswer":{"@type":"Answer","text":"Sending only the most relevant, highest-quality data rather than filling a million-token context window reduces both computational cost and token spend while improving answer quality, since models are now billed per token and relying on the model to sift through excessive context produces lower-quality outcomes than curating precisely what it needs beforehand. Teams optimizing token usage and answer quality in LLM pipelines can keep up with these practices via daily.dev."}}]} ```