Genie Ontology vs Snowflake Semantic Views: what's the difference for AI?

Summary

• A Snowflake semantic view is a schema-level object that stores business concepts as logical tables, relationships, facts, dimensions, and metrics. Snowflake's documentation says Cortex Agents currently reads the definition and generates SQL against the physical tables directly.
• Genie Ontology is the Databricks context layer for the Genie family of AI experiences. It combines modeled Unity Catalog semantics (metric views, domains, and Pages) with snippets that Genie extracts and maintains from your existing assets and usage.
• Both are governed by platform permissions. Genie only uses snippets extracted from assets you have permission to see, and Snowflake documents privileges for creating, replacing, and querying semantic views.
• Both offer AI help with setup. Snowflake's Semantic View Autopilot drafts a semantic view from inputs such as example queries, table metadata, and query history, while Genie Ontology infers snippets from the assets you already have.
• Genie Ontology is in Public Preview, and Unity Catalog Pages, one of its curated inputs, is in Beta.

Genie Ontology vs Snowflake Semantic Views: what's the difference for AI?

A Snowflake semantic view is an object you define. It's a schema-level object that holds logical tables, relationships, facts, dimensions, and metrics, and Cortex Agents, the Cortex Analyst REST API, BI tools, and SQL queries can use it. Genie Ontology is a context layer for the Genie family. It brings together semantics you model in Unity Catalog (metric views, domains, and Pages) with snippets that Genie automatically extracts and maintains from your existing assets and usage. Genie Ontology is in Public Preview.

What does a semantic layer do for AI?

A semantic layer gives an AI system the business meaning of your data. Snowflake's documentation says semantic views address the mismatch between how business users describe data and how it's stored in database schemas, and that they improve accuracy by combining LLM reasoning with rule-based definitions. On Databricks, metric views separate measure definitions from the dimensions used to group, filter, and aggregate them, so you define a metric once and query it at runtime.

How do Genie Ontology and Snowflake Semantic Views compare?

Aspect	Genie Ontology	Snowflake Semantic Views
What it is	The unified context layer that gives the Genie family of AI experiences a business-aware map of your organization	A schema-level object that stores semantic business concepts in the database
Modeled content	Unity Catalog semantics: metric views, domains, and Pages	Logical tables, relationships defined through joins, facts, dimensions, and metrics
Inferred or AI-assisted content	Snippets that Genie extracts and maintains from metric views, dashboards, SQL queries, and Genie Agents	Semantic View Autopilot generates logical tables, relationships, and metrics from inputs such as example queries, table metadata, and query history
How you create it	Define metric views, domains, and Pages in Unity Catalog; Genie also extracts snippets automatically	Semantic Studio in Workspaces, SQL commands such as CREATE SEMANTIC VIEW, or a Snowsight wizard
What uses it	Genie One and Genie Code search the same ontology	Cortex Agents, the Cortex Analyst REST API, BI tools, and SQL queries
Permissions	Snippets are gated by Unity Catalog permissions	Privileges govern creating, replacing, and querying a semantic view
Lifecycle	Public Preview (Pages is in Beta)	Defining semantic views was announced as generally available in June 2025

How does Genie Ontology work?

Genie Ontology brings together two categories of context. Unity Catalog semantics are modeled context that you define, govern, and certify through features like metric views, domains, and Pages. Inferred context is a map of snippets that Genie automatically extracts and maintains from your existing assets and usage, including metric views, dashboards, SQL queries, and Genie Agents.
Each snippet gets an authority score based on where it was generated from, how often it is used, and how fresh it is. When you ask a question, Genie ranks the most relevant snippets, resolves conflicts, and answers using only the sources you are permitted to see. Citation icons in a response show which knowledge sources Genie used. Because permissions decide what context the ontology can retrieve, two people can ask the same question and get different answers based on what each is authorized to see.
Pages give data stewards a governed place to define the authoritative meaning of a concept. Databricks describes the split this way: Pages tell Genie what a concept means, domains tell it where to look, and certification tells it what to trust.

How does a Snowflake semantic view work?

In a semantic view you define logical tables for business entities such as customers or orders and relate them through joins on shared keys. You then add facts (row-level attributes), dimensions (categorical attributes), and metrics (measures calculated by aggregating facts). You can build one in Semantic Studio in Workspaces, with SQL commands such as CREATE SEMANTIC VIEW, or with a wizard in Snowsight. Semantic View Autopilot is the AI-assisted route: you describe what you want, and it generates the logical tables, relationships, and metrics.

FAQs

Is Genie Ontology generally available?

No. Genie Ontology is in Public Preview, and Unity Catalog Pages is in Beta.

How do metric views relate to Genie Ontology?

Metric views are one way you model context for the ontology. They define measures and dimensions once, and Genie Agents, SQL, notebooks, dashboards, and alerts can query them. The ontology also draws on domains, Pages, and the snippets Genie infers.

What does Cortex Agents do with a semantic view?

Snowflake's documentation says Cortex Agents currently reads the information captured in the semantic view definition and generates the SQL against the physical tables directly.

Where does each one live?

Genie Ontology builds on Unity Catalog semantics and respects Unity Catalog permissions, and a semantic view is a schema-level object in a Snowflake database. Each is managed on its own platform.