Knowledge Catalog: Gemini context graph for agents and governance

The overview states that as of April 10, 2026, Dataplex Universal Catalog is called Knowledge Catalog, with API, client library, CLI, and IAM names unchanged. Knowledge Catalog is presented as a Gemini-powered data catalog that supplies business context and governance across the data estate. It automatically extracts semantics from structured and unstructured data into a dynamic context graph used to ground AI agents and reduce hallucinations, and to help data teams discover data, enforce policies, and retrieve context for analytics and autonomous applications.

The rename is explained as a shift from a passive metadata registry to an active, AI-powered context graph as generative AI adoption grows and agents need deep business context. Differences called out include automatic curation of metadata, business logic, and relationships into one graph that agents can treat as enterprise truth for complex tasks, including capabilities such as automatic content-oriented features referenced in the extract.

For ARAMAI-relevant work, the catalog’s claim is shared, machine-usable meaning: semantics lifted into a graph that both governance controls and agents read. That sits at the intersection of data governance and AI agent grounding—one context layer rather than separate dictionaries for humans and models.