Databricks Introduces Unified Trace Table for AI Gateway Monitoring (Beta)
Databricks has launched a unified trace table, now in Beta, designed to centralize the monitoring of AI activity across Unity AI Gateway services. This feature captures all requests and responses in a Unity Catalog table, using OpenTelemetry format. It provides a single, consistent location for monitoring, debugging, securing, and auditing AI operations. The new capability helps engineers streamline their oversight of AI workloads.
Features (1) ›
- Unified Trace Table for AI Gateway Services in Beta
The unified trace table, now in Beta, captures every request and response across Databricks Unity AI Gateway services. This data is stored in a Unity Catalog table in OpenTelemetry format, offering a single location to monitor, debug, secure, and audit all AI activity.
https://docs.databricks.com/aws/en/release-notes/product/2026/august#monitor-all-ai-activity-with-the-unified-trace-table-beta
Related releases
- Databricks Enhances Dashboards and Genie Code with New Features, Fixes Databricks Release Notes ·
- Databricks Genie Agents Enhance Capabilities with GA Agent Mode APIs and Unity Catalog Functions Databricks Release Notes ·
- Python UDTFs in Unity Catalog Now Generally Available Databricks Release Notes ·
- Databricks Enhances Dashboards and Genie Code with AI-Powered Features Databricks Release Notes ·
- Databricks Rolls Out Automatic Upgrades for Row Tracking and Checkpoint V2 Databricks Release Notes ·
- Databricks Unity AI Gateway Nearing General Availability for Enterprise AI Governance Databricks Release Notes ·