November 19, 2026 · Google CloudFeature
BigQuery agent observability generally available, AI indexes in Preview
You can now enable agent observability with the Data Engineering Agent to monitor the agent's token consumption, request volume, and model latency in Cloud Monitoring and inspect execution traces and tool calls in Cloud Trace. This feature is generally available (GA).
You can now create AI indexes using the
CREATE AI INDEX statement
to perform natural language searches across table columns with the
AI.SEARCH function
without manually generating embeddings or managing models. To learn how to
create, monitor, and delete AI indexes, see
Use AI indexes. This feature is in
Preview.
You can use the validation service in the BigQuery Migration Service to verify that tables migrated from your source data warehouse match your target tables in BigQuery across schemas, aggregate metrics, and row-level data. This feature is available in Preview.
BigQuery generative AI functions
now support the gemini-4-neon model. This model is
generally available.
Text from the Google Cloud release notes, licensed under CC BY 4.0. Source: Google Cloud release notes.
Free. Unsubscribe any time.
Related updates
BigQuery Data Engineering Agent Google Drive input and Connected Sheets analytics are GA
BigQuery table-level data lineage generally available, gemini-4-hydrogen in AI functions
BigQuery compute cost attribution in Preview, Zstd compression functions in Preview
Generative AI observability now generally available for Amazon CloudWatch · AWS
Amazon CloudWatch adds generative AI observability (Preview) · AWS