Serverless compute access control is now generally available

Serverless compute access control is now generally available. Workspace admins can govern who runs serverless workloads through two built in serverless compute objects: Default Interactive Compute (governs access to notebooks and Databricks Connect) and Default Automated Compute (governs access to jobs and Spark Declarative Pipelines on Lakeflow). All workspace users have Can Use on both objects by default, so existing workloads continue to run without changes.