What Is Groundedness in AI? (Citations Explained)

Microsoft Foundry IQ lets AI answer from YOUR documents, with citations. Here's why groundedness is the word every enterprise AI project will ask you about. This clip is from a live session in my Microsoft Foundry series. A hospital doesn't want generic ChatGPT answers. It wants ChatGPT-level intelligence answering from its own data, with sources. Foundry IQ takes care of chunking, embeddings and retrieval for you. I show where to add a knowledge base, and why wrong retrieval gives wrong answers. What you'll learn: - Why enterprises want AI grounded in their own knowledge base - What Foundry IQ does for you: chunking, embedding and retrieval - Citations and groundedness, and why every AI project asks for them - Custom GPTs vs an enterprise knowledge base with years of data - Adding a knowledge base in Foundry: files, Blob Storage, web, MCP sources - Why AI gives wrong answers: retrieving the 2024 HR policy instead of 2026 Chapters: 0:00 Why Enterprises Want Their Own Data 0:37 Chunking & Embedding, Done by Foundry IQ 1:49 Citations: Answers with Sources 2:38 Groundedness Explained 3:01 Custom GPT vs Enterprise Knowledge 4:09 Add a Knowledge Base in Foundry 5:10 Why Knowledge Sources Matter 5:43 Wrong Retrieval = Wrong Answer Watch next: ▶ Microsoft Foundry - AI Platform (full playlist): https://www.youtube.com/playlist?list=PLeGiFBPpRC04 Connect: 🎓 Courses: https://gauravkhurana.com/courses 🌐 Website: https://gauravkhurana.com 📅 1:1 mentoring: https://topmate.io/gauravkhurana #MicrosoftFoundry #FoundryIQ #Groundedness #RAG #SharingIsCaring