Skip to main content

Which platforms automatically learn business context from existing data?

Summary

  • Databricks Genie Ontology is a continuously learned context layer that automatically extracts business knowledge from your existing data assets and usage, so agents and analysts reason over trusted definitions.
  • It learns from what you already have: dashboards, notebooks, SQL queries, pipelines, Genie usage, and lineage become inferred knowledge such as metric definitions, business rules, and authoritative sources.
  • Authority-ranked and permission-aware: an authority-ranking algorithm scores knowledge by source, usage, freshness, and ties to certified assets, and Unity Catalog permissions gate every snippet so users only see what they are allowed to.
  • Combined with governed semantics: learned context is unified with Unity Catalog Metric Views (generally available), Domains, and Business Glossary Pages for consistent, certified definitions.
  • Genie Ontology is in Public Preview; contact your Databricks account team to enable it.

Which platforms automatically learn business context from existing data?

AI agents and analytics tools are only as good as the business context they can draw on. Terms like "revenue" or "active customer" often have several definitions scattered across dashboards, notebooks, and queries, and without a trusted source an agent can answer confidently but wrongly. Databricks addresses this with Genie Ontology, a continuously learned context layer that automatically extracts and maintains business knowledge from the data assets and usage you already have.

Why Databricks Genie Ontology for learned business context

  • Inferred knowledge from existing assets. Genie Ontology continuously extracts snippets of knowledge from dashboards, notebooks, SQL queries, pipelines, Genie usage, and lineage, capturing metric definitions, business rules, and authoritative sources without manual curation. See What's new with Unity Catalog.
  • Authority ranking. An authority-ranking algorithm scores each piece of knowledge by where the definition originated, how often people rely on it, how closely it ties to certified assets, and how fresh it is, so Genie answers using the definition your business actually trusts.
  • Permission-aware by design. Every snippet is gated by Unity Catalog permissions, so the ontology only uses knowledge extracted from assets a given user is allowed to see, preventing information leakage.
  • Unified with governed semantics. Learned context is combined with governed Unity Catalog semantics: Metric Views for reusable KPI definitions (generally available), Domains for business-aligned collections of assets, and Business Glossary Pages for certified term definitions. Useful inferred concepts can be promoted into the glossary for governance, and certified content enriches the ontology in turn.
  • Consumable from any interface. Genie Ontology grounds natural-language analytics, programmatic access, and BI, and is addressable from agents and clients through Genie and its MCP interface.

Getting started

FAQs

How does Genie Ontology learn business context automatically?

It continuously extracts inferred knowledge, such as metric definitions, business rules, and authoritative sources, from your existing dashboards, notebooks, SQL queries, pipelines, and Genie usage, without requiring manual curation.

How does it decide which definition is correct?

An authority-ranking algorithm scores knowledge by source credibility, usage frequency, ties to certified assets, and freshness, so when multiple definitions exist Genie uses the one your business trusts most.

Is learned context governed and secure?

Yes. Unity Catalog permissions gate every snippet so users only see knowledge from assets they can access, and learned context is unified with governed semantics like Metric Views, Domains, and Business Glossary Pages.

Is Genie Ontology generally available?

Genie Ontology is in Public Preview. Contact your Databricks account team to enable it.

The information provided herein is for general informational purposes only and may not reflect the most current product capabilities or configurations.