Battle for AI Data Platform

This rivalry matters because the winning platform is no longer just a place to store and query data, it is becoming the operating system for how enterprises build reports, train models, and ship AI apps. Databricks moved down into SQL and dashboards with Databricks SQL, while Snowflake moved up into developer and ML workflows with Snowpark and Cortex. That pulls both vendors into the same budget, the same buyer, and the same daily workflow.1 2 3 4 5 6

• Databricks started from data engineering and ML, but Databricks SQL turned it into a real warehouse competitor. By mid 2024, Databricks SQL had reached about $400M in annualized revenue, up 300% YoY, showing that BI and analyst workloads were no longer just an add on.1 3 4
• Snowflake started as the easier warehouse for analysts, then added developer and ML tooling so teams could stay inside Snowflake for Python, feature work, model training, and AI app building. Snowpark gives a Python and Spark like interface, and Cortex pushes further into AI products.5 6 7
• The overlap changes adjacent vendors too. dbt is under pressure as both Databricks and Snowflake build native transformation features, and governance vendors like Immuta now treat both platforms as the same control point because large enterprises often run both side by side.2 8

The next phase is a fight to own the full AI data loop, from raw data and SQL analysis to model training, governance, and natural language interfaces for business users. That favors the platform that can make more workloads feel native in one place, while keeping costs low enough that customers stop stitching together separate tools.2 4 6 7 8