Updates to conversational analytics include the following improvements:

Updates to conversational analytics include the following improvements:

• ObjectRef support: BigQuery conversational analytics now integrates with Google Cloud Storage through ObjectRef functions. This lets you reference and interact with unstructured data such as images and PDFs in Cloud Storage buckets in your conversational analysis. • BQML support: BigQuery conversational analytics now supports a set of BigQuery ML functions, including AI.FORECAST, AI.DETECT_ANOMALIES, and AI.GENERATE. These functions let you perform advanced analytics tasks with simple conversational prompts. • Chat with BigQuery results: You can now start conversations and chat with query results in BigQuery Studio (SQL editor). • Enhanced support for partitioned tables: BigQuery conversational analytics can now use BigQuery table partitioning. The agent can optimize SQL queries by using partitioned columns such as date ranges on a date-partitioned table. This can improve query performance and reduce costs. • Labels for agent-generated queries: BigQuery jobs initiated by the conversational analytics agent are now labeled in BigQuery Job History in the Google Cloud Console. You can identify, filter, and analyze the jobs run by the conversational analytics agent by referencing labels similar to {‘ca-bq-job’: ‘true’}. These labels can help with the following tasks:

Monitor and attribute cost. • Audit agent activity. • Analyze agent-generated query performance.

This feature is available in Preview.