Cloud data storage giant Snowflake has solidified its strategic alliance with Amazon Web Services (AWS) by signing a significant $6 billion, five-year agreement. While Snowflake has historically operated on AWS since its inception in 2012, and now also supports Microsoft Azure and Google Cloud, this new contract underscores a substantial commitment. AWS reported that Snowflake has generated a total of $7 billion in services via the AWS Marketplace since 2012, making this latest deal nearly equivalent to all previous earnings from that cloud platform.

The impetus behind this accelerated investment is the burgeoning field of Artificial Intelligence (AI). Snowflake indicates that its customers are dramatically increasing their spending on AWS for AI initiatives, with projections showing this spend doubling to $2 billion in the calendar year 2025 alone. Snowflake has been proactively addressing this demand through its AI building tool, Cortex AI, which has been available for several years. Cortex AI leverages Snowflake's position as a central repository for enterprise data, offering features such as natural language text interfaces for database queries and automated summary reports, thereby making AI capabilities more accessible to businesses.

A critical component of this new agreement is Snowflake's expanded access to AWS's proprietary ARM-based CPU chip, Graviton. The shift of AI from intensive training phases to widespread daily usage and automation via AI agents has led to an exponential increase in demand for CPU processing power. While Graphics Processing Units (GPUs) are primarily responsible for AI training and complex reasoning tasks, Central Processing Units (CPUs) manage the majority of other AI-related operations, particularly those involving agents. Amazon CEO Andy Jassy previously highlighted Graviton's "better price-performance" compared to Nvidia's offerings, even as AWS continues to deploy Nvidia's chips across its cloud infrastructure.

The intense demand for AI processing capabilities is driving cloud providers like AWS to rapidly deploy new chip technologies. Amazon, known for its cost-efficiency, passes the savings from its homegrown chips on to its customers, which in turn is attracting major multi-billion-dollar deals. For instance, AWS recently secured a deal to supply millions of Graviton chips to Meta to support its expanding AI compute requirements. This was a notable win for AWS, especially considering Meta's earlier $10 billion agreement with Google Cloud.

These developments signify a growing trend among cloud giants to develop their own AI chips, posing a direct challenge to established chipmakers like Nvidia. Google has been developing its own AI chips for years, and Microsoft recently unveiled its Maia AI chip in January. Unsurprisingly, Nvidia CEO Jensen Huang has expressed a strong resolve to defend and expand his company's market dominance. Last week, following another record-breaking financial quarter, Huang announced Nvidia's new AI-specific CPU, Vera, declaring it represents a "brand new" $200 billion market for the company and claiming $20 billion in sales already.

While Nvidia is unlikely to cede market share easily to Amazon or other cloud providers, the multi-billion-dollar cloud deals underscore how AI is profoundly impacting and benefiting the cloud infrastructure sector. Regardless of which specific companies ultimately gain the most from the pervasive integration of AI into our professional and personal lives, cloud providers are clearly securing a substantial share of this burgeoning market.