In more good news for Amazon, Snowflake signs $6B deal with AWS for AI CPU chips

Snowflake just committed $6 billion to Amazon Web Services over the next five years, specifically for AI CPU chips. This isn't a general cloud spend increase. It's a targeted bet on AWS's custom silicon for AI workloads.

The deal is a big statement about where AI infrastructure is heading. Nvidia has dominated the AI chip market, but cloud providers like AWS are investing heavily in their own alternatives. Snowflake's willingness to commit this much capital suggests those alternatives are becoming viable for production workloads.

For AWS, this is validation that their chip strategy is working. They've been building custom processors like Trainium and Inferentia to offer customers cheaper, more efficient options than Nvidia GPUs. A $6 billion vote of confidence from a major data platform shows those chips are ready for serious AI work.

If you're running AI workloads in the cloud, this matters because it signals more competition in the chip market. More competition typically means better pricing and more options. It also means the assumption that you need Nvidia GPUs for AI is becoming less universal.

Snowflake's move also reflects a broader trend of companies locking in long-term infrastructure deals as AI compute becomes a strategic resource. When you're betting your product roadmap on AI capabilities, securing reliable access to chips at known prices makes sense, even if it means a massive upfront commitment.