The Pulse
Trust3 AI Syncs Databricks and Snowflake Policies to OneLake
Trust3 AI announced a service to synchronize row- and column-level access policies from Databricks and Snowflake with Microsoft OneLake Security. The integration is designed to carry existing permissions into Microsoft Fabric and Power BI a
AI.info Team · Written with AI ·
On September 29, Trust3 AI announced a service that carries row- and column-level security policies from Databricks and Snowflake into Microsoft OneLake Security. The aim is to preserve existing access decisions when enterprise data becomes available through Microsoft Fabric and Power BI, including when autonomous agents retrieve it. Trust3 says the service synchronizes policy changes automatically rather than requiring teams to recreate permissions across platforms.
One dataset, more than one access route
OneLake can make data held in systems such as Databricks and Snowflake accessible through Microsoft’s analytics environment. That creates a governance problem: permissions set in the source system do not automatically become equivalent OneLake permissions. Trust3’s own technical explanation says teams otherwise have to map identities and recreate permissions in Fabric, with policy changes potentially leaving the two environments out of step.
The company says its integration translates source-platform rules into OneLake Security rules and keeps them aligned as access changes. Its technical post describes mapping users and groups into OneLake roles and resolving details such as the tenant, workspace, lakehouse and resource. Trust3 says the source platform remains the authority for the original policy; OneLake receives a translated version for access through Microsoft’s environment.
Agents inherit the same access boundaries
The announcement frames the integration partly around AI agents that may select different ways to reach data. If permissions vary by access path, an agent using Fabric or Power BI could encounter different controls from a person or system querying the original platform. Trust3 says its synchronization is meant to carry those controls across both routes, including for Power BI and Fabric Data Agents.
“Enterprise AI agents should inherit the same trust boundaries as the people and systems they act on behalf of,” Don Bosco Durai, Trust3 AI’s CTO, said in the announcement. He said the integration carries security intent from Databricks and Snowflake into an agent’s identity and permissions in Microsoft Fabric, without creating a separate security model.
Dipti Borkar, Microsoft’s vice president of Microsoft IQ and OneLake, said access controls need to continue across OneLake, Databricks and Snowflake as companies adopt data from multiple platforms. Her statement backs the cross-platform governance goal, but the announcement does not describe a Microsoft product launch or provide independent test results for Trust3’s integration.
What the announcement leaves unspecified
The release does not give pricing, a customer deployment example, performance measurements or a detailed technical account of how quickly policy changes appear in OneLake. It also does not describe how conflicts or synchronization failures are handled. Those details matter to security teams assessing whether translated permissions match the original policy in every supported configuration.
Trust3 published a technical explanation of its OneLake approach on August 7, describing synchronization and identity mapping for Databricks and Snowflake data surfaced through Fabric. The new announcement extends that message to autonomous agents and specifically names row- and column-level policies. For organizations using those platforms together, the practical test is whether a changed permission produces the same access decision through the source system and through OneLake.