AI BriefingGoogle Cloud launches Data Cloud connectors in Gemini for conversational analytics
·2026.11.06 09:00
Key point
The preview feature supports BigQuery, Spanner, AlloyDB, Cloud SQL, and Looker, using Knowledge Catalog to enforce row- and column-level security.
Details
- Data Cloud connectors in Gemini for business are now in preview, allowing users to query live enterprise data in natural language without writing SQL.
- The system uses Knowledge Catalog to interpret questions, map business terms to database structures, and execute permission-scoped queries under the user's identity.
- Supported data sources include BigQuery, Spanner, AlloyDB, Cloud SQL, and Looker, with the BigQuery connector enabling cross-cloud queries of open table formats like Apache Iceberg, Delta Lake, and Parquet in AWS or Azure.
- Connectors use federated query mode to access data in place without copying or indexing it, distinguishing them from data ingestion modes that copy data into a Gemini store.
- Two usage paths are available: direct Data Cloud connectors for ad hoc analysis and Conversational Analytics agents for scoped, certified multi-step analyses built by data teams.
- First-party connectors offer built-in skills for higher response quality and managed authentication, removing the need for self-managed Model Context Protocol (MCP) middleware or OAuth configuration.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.