Introducing OpenSharing: the Next Evolution of Delta Sharing for the Agentic Era

Summary

OpenSharing is presented as the next evolution of Delta Sharing, extending an open zero-copy data-sharing protocol from tables and files to models, agents, semantic context, unstructured data, and reusable AI logic. The protocol is now an independent open-source project hosted by the Linux Foundation, while Databricks OpenSharing adds Unity Catalog governance and audit logging, Marketplace discoverability, and enterprise features. Genie Agent Sharing supports governed AI experiences across organizational boundaries, with controls for proprietary instructions, data access, daily prompt quotas, and row exports. SecureConnect removes per-recipient firewall changes through a Databricks-managed proxy, while Global Distribution uses local replicas to reduce egress fees and latency. The launch also supports Apache Iceberg REST Catalog API, external catalogs, and on-premises storage partners; providers define shares in Unity Catalog, recipients query live data through existing tools, and governance enforces access controls.

Context

Delta Sharing was designed for sharing tables and files, but the source says organizations now need to exchange semantic context, AI skills, unstructured data, and autonomous agents across cloud, vendor, and company boundaries. Existing protocols are described as vendor-specific, unable to handle AI logic, and dependent on networking setup that can take weeks for each new partner.

Approach / What changed

OpenSharing extends Delta Sharing as an open-source protocol for the full AI stack, supporting data, models, agents, and Genie Agents across formats, clouds, and organizational boundaries. Databricks provides an enterprise implementation using Unity Catalog for governance and audit logging, Marketplace for discoverability, SecureConnect for managed networking, Global Distribution for replication, and interoperability with Iceberg clients and on-premises storage ecosystems.

Takeaways

• OpenSharing natively supports Delta Lake, Apache Iceberg, and Parquet, allowing data and AI assets to be shared across clouds, vendors, and organizational boundaries.
• Genie Agent Sharing lets providers share semantic context, business metrics, and reusable AI logic while controlling proprietary instructions, data access, daily prompt quotas, and row export limits.
• SecureConnect uses a Databricks-managed proxy to avoid per-recipient firewall changes, while Global Distribution provides local replicas intended to reduce cross-cloud egress fees and latency.

Related reading