Databricks has shared new platform security and compliance updates at Data + AI Summit 2026 to help organizations scale Genie, serverless, Lakebase, and AI workloads with stronger identity controls, private connectivity, and broader compliance coverage across clouds.
What’s new
- Automatic Identity Management is expanding: AIM for Microsoft Entra ID is now available on AWS, and AIM for Okta is in Public Preview, helping teams provision and manage users, groups, and service principals more directly from their identity provider.
- More context-aware access controls: Context-based ingress is in Public Preview on AWS and lets admins define access rules based on identity, network source, and request type, including for workspace UI, APIs, Apps, and Lakebase compute.
- Private connectivity for performance-heavy services: Databricks has expanded inbound private connectivity for services such as Lakebase Autoscaling and Zerobus Ingest through service-direct PrivateLink / Private Link on AWS and Azure.
- Broader compliance support across clouds: Databricks now lists HITRUST support across AWS, Azure, and GCP, and ISMAP coverage for Japan’s government-focused cloud security framework.
- More regulated-cloud support for AI and analytics: AWS GovCloud release notes now include features such as Databricks Apps, AI Search, Predictive Optimization, Genie, and Genie Code, while FedRAMP High is also listed for Azure Commercial through Databricks’ trust resources.
👉 Read the full post here
