What happened
Snowflake and Google Cloud have aligned their data platforms around Apache Iceberg, an open table format that lets multiple engines (BigQuery, Apache Spark, Trino, Snowflake) read and write the same data without copying it. Snowflake's Horizon Catalog and Google Cloud's Lakehouse runtime catalog enable bidirectional federation—each platform's engines can access tables managed by the other through standardized REST API endpoints, with vended credentials controlling access.
Why it matters
Organizations no longer face a choice between flexibility (letting teams use different tools but risking data inconsistency) and consistency (forcing one engine but losing specialization). By anchoring both platforms to Iceberg and shared catalog governance, teams can pick the best tool for each job while central policies enforce security, access control and unified semantics. This matters especially for AI: Snowflake exposes model context protocol (MCP) servers so AI agents (Gemini Enterprise, code assistants) can query lakehouse data through the same governed channels humans use, while semantic models stored in the data layer (via Semantic View Autopilot and Semantic Studio) reduce AI hallucination by embedding verified business logic.
What to watch
Snowflake Horizon Context and Google Cloud's Knowledge Catalog embed business context (table descriptions, data quality signals, usage patterns) directly in the catalog so AI agents receive not just data but the surrounding knowledge needed to interpret it correctly—a distinction the body illustrates with the example of a grounded response ('Q4 revenue was $47.2M from the authoritative finance.revenue table') versus a generic one ('revenue might be...'). Semantic Studio's CoCo assistant is in private preview.
Summaries like this, in your inbox every morning.
The article describes a convergence of two technical trends: interoperability and AI-readiness. For years, organizations faced a painful tradeoff—either let each team use its preferred engine (Apache Spark, BigQuery, Snowflake) and accept data fragmentation, or force everyone through a single system and lose specialization. Apache Iceberg, adopted across the industry, solved the first problem by establishing a shared physical layout so all engines can read and write the same tables without proprietary adapters. Snowflake and Google Cloud have now gone further by building managed catalogs (Horizon and Lakehouse runtime catalog) that federate across platform boundaries, using standardized REST APIs and short-lived, narrowly scoped storage tokens to enforce governance regardless of which engine accesses the data.
The second innovation addresses AI. Simply making data accessible is not enough; AI systems hallucinate when they lack semantic grounding—the business logic, definitions and context needed to interpret raw data correctly. Snowflake's Semantic View Autopilot and Semantic Studio (with a CoCo assistant in private preview) automatically discover and maintain semantic models from query patterns and user behavior, embedding business logic in the data layer rather than leaving it to AI prompts. Google Cloud's Knowledge Catalog provides similar contextual metadata through Gemini-powered insights. Both platforms expose these models and context to AI agents through model context protocol servers and REST APIs, ensuring agents inherit the same governance and semantic grounding as human analysts.
Pick your industry and the AI tools you use, and get news related to your work every day.
Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.
Chunghwa Telecom president Rong-shy Lin said the shift from conversational AI to autonomous AI agents is resha…
CoreWeave fell 5% in premarket trading Friday and was on track to finish the week down 15%, after rebounding f…
Apple and Meta said they would jointly release a new open-weight AI model designed to run on high-end Macs, ai…
Investors put $38.5 billion into Asian startups in Q3, per Crunchbase data
A writer and entrepreneur opened his daily to-do list and found 156 items, and concluded that an AI assistant…
A longtime Anthropic investor told Fortune he expects shares priced around $1.8 trillion, below the company's…
