Pages in Unity Catalog: Govern Business Knowledge for Humans and AI

When AI agents don't know what your business terms actually mean, they guess — and your teams stop trusting the answers. Pages, built into Databricks' Unity Catalog, gives data stewards a native home to author and publish authoritative definitions of business concepts: metrics, terms, entities, and more. In this demo, you'll see how Pages works end-to-end: How Pages fits into Genie Ontology and Unity Catalog Semantics Authoring a Page with Genie Code (AI-assisted drafting from your existing source material) How business users discover and search authoritative definitions in Discover How Genie One grounds its answers in Pages — with citations Collaboration and governance workflows for keeping definitions accurate Bulk bootstrapping: turning an existing glossary spreadsheet into Pages in minutes Whether you're a data steward governing definitions, a business analyst trying to trust your metrics, or a platform admin bringing order to a sprawling data estate — Pages closes the loop between your data and the business context that makes it meaningful. 0:00 The AI Context Problem 0:45 Introducing Genie Ontology 1:45 What is Pages? 2:20 Navigating Pages in Unity Catalog 3:10 Creating a Page: Structured Fields & Synonyms 4:15 AI-Assisted Authoring with Genie 5:10 Enriching Pages & Linking Data Assets 5:50 Draft & Publish Workflow 6:25 Business User: Finding Authoritative Answers 7:20 Genie Q&A Grounded in Ontology 8:40 Synonym Matching & Citation in Action 9:10 Collaboration: Suggestions & Comments 9:55 Bulk Bootstrapping from Existing Glossaries 11:00 Summary & Beta Access Explore how to set up domains, author Pages, and connect them to Genie: https://docs.databricks.com/aws/en/uc-semantics/pages