Google Cloud Brings LookML Semantics to Knowledge Catalog
LookML semantic indexing is now generally available for Looker Core on Google Cloud, feeding business definitions to AI agents.
Updated on Nov. 10, 2026 in Artificial Intelligence

Google Cloud has made LookML semantic indexing in Knowledge Catalog generally available for Looker Core on Google Cloud. The release indexes resolved LookML models, Explores, Views, lineage and dashboards so AI agents can draw on defined business logic.
Why it matters
Data catalogues describe what data exists and what governance applies but do not define how figures should be calculated in SQL. Without explicit semantic rules, AI agents may assume the wrong fiscal calendar, include transactions that should be excluded, or expose restricted data fields.
The indexing captures join logic, relationship cardinality, required joins and mandatory filters at the Explore level, with production LookML changes reflected in near real time. The article gives no hard performance or accuracy metrics for the feature.
The players
Google Cloud
Google Cloud is the cloud computing division of Alphabet Inc. that offers data analytics, AI and enterprise infrastructure products.
Looker
Looker is a business intelligence and data analytics platform, acquired by Google, that uses LookML as its modelling language.
The details
An AI agent first searches Knowledge Catalog for governance information and relevant LookML objects, then generates SQL grounded in those indexed definitions. Access controls remain tied to end-user credentials, and users can attach a BigQuery, Spanner, Cloud SQL or AlloyDB data store in federated query mode to enable Knowledge Catalog support automatically.
Timeline
November 10, 2026: Google Cloud announces general availability of LookML indexing for Looker Core on Google Cloud.
The Tech Race
The move extends Looker's LookML modelling language, long used to define joins and metric formulas for dashboards, into the layer that AI agents read before writing SQL. It positions Google's data stack to compete with rival platforms that are also trying to give enterprise agents governed, business-specific context.
Analysts and data teams using Looker Core on Google Cloud can have AI agents draw on existing LookML definitions rather than rebuilding business logic elsewhere. The indexing syncs by default, so changes to production LookML reach the catalog in near real time.
The takeaway
Semantic definitions such as fiscal calendars and exclusion rules are what keep AI-generated SQL aligned with how a business actually counts its numbers. Keeping those rules in the same place as the data governance layer makes them easier to audit and reuse across agents.
Further reading
For more on how AI systems are being grounded in enterprise data, explore Artificial Intelligence coverage.










