OSI: Realizing semantic layer portability

Define a metric in one tool. Redefine it in the next. Watch your dashboards, notebooks, and AI agents disagree on what "active customer" means: each one is reading a different definition. Semantic fragmentation is the quiet tax every data team pays, and it's the reason AI outputs still can't be trusted at scale.

The Open Semantic Interchange (OSI) changes that contract. Co-led by dbt Labs, Snowflake, Salesforce, BlackRock, and RelationalAI, OSI is a vendor-neutral, open-source specification for how semantic definitions move between tools. Author your metrics once in the dbt Semantic Layer, and those definitions become portable to the OSI partner ecosystem.

You'll see how MetricFlow, now open-sourced under Apache 2.0, compiles governed metrics into the OSI interchange format, how the dbt Semantic Layer exposes those definitions to downstream tools, and how teams are using this today to bring conversational analytics, BI, and AI agents into agreement on the same business logic.

You'll leave with a clear picture of what OSI unlocks for your stack, the migration path from tool-specific semantics to portable ones, and a framework for deciding which metrics to standardize first.