Snowflake has announced a raft of new product innovations aimed at helping enterprises move data and AI initiatives out of experimentation mode and into full-scale production more quickly. The updates are designed to simplify development, improve interoperability and ensure AI systems operate on trusted, governed enterprise data.
At the centre of the announcement is the general availability of Cortex Code, a data-native AI coding agent built to automate and accelerate enterprise development end to end. Unlike generic coding assistants, Cortex Code is designed to understand an organisation’s Snowflake data environment, including governance rules, compute usage and operational semantics, allowing teams to build pipelines, analytics and AI applications faster without compromising security or trust.
Snowflake has also made Semantic View Autopilot generally available. The AI-powered service automates the creation, optimisation and ongoing governance of semantic views, giving AI agents a consistent and shared understanding of business metrics. By reducing manual semantic modelling, which can take days, Snowflake says organisations can cut this down to minutes while lowering the risk of inconsistent definitions and AI hallucinations.
In parallel, Snowflake is extending its database strategy with enhancements to Snowflake Postgres, which will run natively within the AI Data Cloud. By bringing transactional, analytical and AI workloads onto a single platform, Snowflake aims to eliminate the complex pipelines that typically connect operational and analytical systems, reducing cost, risk and development time.
Christian Kleinerman, Executive Vice President of Product at Snowflake, said that enterprises are increasingly under pressure to deliver tangible value from AI rather than isolated proofs of concept. According to him, embedding AI directly into the development lifecycle and making data “AI-ready by design” represents a fundamental shift in how organisations build with data and AI at enterprise scale.
Deepening ties with OpenAI
Alongside the product announcements, Snowflake highlighted a new multi-year, $200 million partnership with OpenAI. The collaboration focuses on co-innovation and joint go-to-market efforts, and makes OpenAI models natively available to Snowflake’s 12,600 global customers through Snowflake Cortex AI.
As part of the agreement, OpenAI models, including GPT-5.2, will also be accessible through Snowflake Intelligence, the company’s enterprise intelligence agent that allows employees to query organisational data using natural language. Customers such as Canva and Whoop are already using the integration to run OpenAI models directly on their proprietary enterprise data.
Reducing friction in enterprise AI development
Snowflake positioned Cortex Code as a response to growing frustration among teams that need to move faster while maintaining accuracy, governance and scale. The agent works across the full development lifecycle, from design and implementation to optimisation and operations, and can be used both within Snowflake’s interface and in local developer environments such as VS Code or Cursor.
To further streamline AI application development, Snowflake is introducing new capabilities for so-called “vibe coding” through an upcoming integration with v0 by Vercel. This will allow users to build AI-powered applications using natural language and deploy them securely within Snowflake using Snowpark Container Services.
Making enterprise data AI-ready by default
Snowflake’s updates to Semantic View Autopilot and Snowflake Postgres are aimed squarely at one of the biggest blockers to AI adoption: fragmented data and inconsistent business logic. Many enterprises still define metrics manually across different systems, leaving AI agents without a shared context and making outputs harder to trust.
By automating semantic governance and consolidating transactional and analytical data, Snowflake argues it is laying the groundwork for AI systems that are reliable, auditable and ready to operate at scale. The company said these moves build on its broader push for open data interoperability, including initiatives such as Open Semantic Interchange, while adding intelligence to continuously maintain business context.
Taken together, the announcements signal Snowflake’s ambition to position its AI Data Cloud as a production-ready foundation for enterprise AI—one that allows organisations to move faster without losing control over data, governance or outcomes.