Anthropic leans on AWS platform

Anthropic just made a very specific enterprise move — and it runs straight through Amazon. On May 11, Anthropic and AWS put Claude Platform on AWS into general availability, which means companies can now use Anthropic’s native platform through their AWS account instead of setting up a separate vendor relationship. That sounds like procurement plumbing. But for big companies, procurement plumbing is the product. It decides who can buy, how security works, where logs go, and whether the spend counts against giant cloud commitments. ### What actually launched? This is not just Claude models inside Amazon Bedrock. It is Anthropic’s own platform experience — the native console, developer tools, and access path — sold through AWS. Anthropic says customers get AWS authentication, billing, and commitment retirement, while still using the full Claude Platform feature set. AWS says it is the first cloud provider to offer Anthropic’s native platform this way. (aws.amazon.com) ### How is that different from Bedrock? Bedrock is AWS’s managed model layer. You call Claude there alongside other model families, and AWS wraps the experience with its own guardrails, knowledge bases, agents, regional controls, and data-handling model. Claude Platform on AWS is the opposite emphasis — you go closer to Anthropic’s own product surface, including early access to beta features and native developer workflows. Basically, AWS is now offering two doors into Claude: the AWS-shaped one and the Anthropic-shaped one. (aws.amazon.com) ### Who runs the service? This is the part governance teams will stare at. AWS says Claude Platform on AWS is operated by Anthropic, and customer data is processed outside the AWS security boundary. But AWS still provides IAM-based access, consolidated billing, and CloudTrail audit logging. So the control plane feels AWS-native, while the underlying service remains Anthropic’s. That split is useful, but it is not the same as “everything stays fully inside AWS” in the Bedrock sense. (aws.amazon.com) ### Why do cloud commitments matter so much? Because large enterprises already prepay or commit huge amounts to AWS. Anthropic’s AWS Marketplace listing for Claude Enterprise says the spend can draw down directly from AWS EDP or PPA commitments. That changes the internal sales motion. Instead of asking a company to add a brand-new AI vendor and budget line, Anthropic can ride an existing AWS relationship. Turns out that can matter more than model benchmarks when a purchase has to clear finance, security, and procurement. (aws.amazon.com) ### Is Anthropic becoming AWS-exclusive? Not exactly. Anthropic said just weeks ago that Claude remains available across AWS, Google Cloud, and Microsoft’s platform offerings. Claude Opus 4.7 is also listed across Anthropic’s native platform, Bedrock, Google Vertex AI, and Microsoft Foundry. So this is not exclusivity. It is deeper commercial alignment with AWS — especially for enterprise distribution. ### Why do partners care? (aws.amazon.com) Because a lot of enterprise AI work is really systems integration work. AWS-focused consultancies are already building dedicated Claude practices, which suggests they see demand from customers who want Anthropic tools but prefer to stay inside AWS procurement and identity rails. Once that pattern sets, partner ecosystems tend to reinforce it. ### What is the catch? (anthropic.com) The catch is choice now comes with a more explicit tradeoff. Bedrock is better if a company wants AWS-managed controls, regional residency, and a cleaner “stay in AWS” story. Claude Platform on AWS is better if a team wants Anthropic’s newest native features and workflows, but can live with Anthropic operating the service. Same model family — different operating model. ### Bottom line? Anthropic is not just selling a model anymore. (crn.com) It is selling a route through the enterprise. And right now, the fastest route looks a lot like AWS. (aws.amazon.com 1) (aws.amazon.com 2)