Snowflake & OpenAI Ink $200M Agent Deal
Snowflake and OpenAI have struck a $200M partnership that will unlock AI agent capabilities for over 12,600 enterprise customers. The deal aims to make it easier for companies and their agencies to build, deploy, and manage custom AI agents for tasks like data analysis and campaign management directly within their existing data infrastructure.
The multi-year, $200 million deal makes OpenAI a primary model provider within Snowflake's ecosystem, integrating models like GPT-5.2 directly into the Snowflake AI Data Cloud. This allows joint go-to-market strategies and co-innovation aimed at deploying customized AI solutions for enterprise clients. This integration is centered on Snowflake Cortex AI, a fully managed service that allows large language models to operate directly where data is stored. This eliminates the need to move sensitive enterprise data to external APIs, a major hurdle for AI adoption due to security and governance concerns. For developers and analysts, the partnership enables calling OpenAI's models directly from SQL commands within Snowflake. According to Snowflake CEO Sridhar Ramaswamy, this allows organizations to build "powerful, responsible, and trustworthy" AI agents on top of their most valuable asset—their own proprietary data. This move is part of a broader enterprise offensive by OpenAI, which recently unveiled its "Frontier" platform for building and managing AI agents. To accelerate adoption, OpenAI has also formed "Frontier Alliances" with major consulting firms like Accenture, McKinsey, and Boston Consulting Group to help large companies deploy its technology. The collaboration deepens an existing relationship, as OpenAI already uses Snowflake for its own internal analytics and experiment tracking, while Snowflake uses ChatGPT Enterprise to support its employees. Early adopters expected to leverage the new integration include global brands like Canva and WHOOP. This partnership signals a strategic evolution for the data warehouse, transforming it from a passive data repository into an active orchestration layer for AI. The move intensifies competition with rivals like Databricks and other cloud providers who are also embedding AI models directly into their platforms.
