Amazon to invest another $25 billion in Anthropic and lock in a 10-year AWS contract

Amazon announced an investment of up to $25 billion in Anthropic — the creators of the Claude language model — in exchange for a 10-year agreement to use AWS cloud infrastructure. The amount represents one of the largest corporate bets in the generative AI market to date, cementing AWS's status as the primary cloud partner of one of the world's fastest-growing AI startups. The new deal marks the second major mega-investment by Amazon in a leading AI lab in two months.

In February 2026, the company invested approximately $50 billion in OpenAI — a structurally similar agreement involving long-term commitment to cloud infrastructure in exchange for financing. Amazon has now become an anchor investor in both leading players in the race for frontier language models — and this is no accident, but rather a deliberate strategy. Under the partnership terms, Anthropic commits to using AWS Trainium chips — including the next-generation Trainium4 — as its primary computing base.

Amazon in turn guarantees the company access to computational capacity of up to 5 gigawatts. To put this in scale: this is comparable to the power output of five large thermal power plants — the exact amount of energy required by modern industrial-scale AI clusters for training and inference of next-generation models. Anthropic is currently experiencing rapid growth.

The company's annual revenue has reached $30 billion — a figure that would have seemed fantastic just a year and a half ago. Claude is actively used by corporate clients through Amazon Bedrock and by developers directly through Anthropic's API. Unlike OpenAI, which bet on mass consumers through ChatGPT, Anthropic focuses on the enterprise segment and positions itself as a responsible developer of AI systems — with emphasis on safety and model interpretability.

For Amazon, this deal makes strategic sense on multiple levels. First, it strengthens AWS's competitive position against Azure — the primary cloud partner of OpenAI and Microsoft — in the battle for enterprise demand for AI infrastructure. Contracts for training and hosting large language models are becoming one of the fastest-growing segments of the cloud market.

Second, tying Anthropic to Trainium chips stimulates scaling of Amazon's own semiconductor division, which directly competes with the dominant NVIDIA. Every petaflop of computation on Trainium is a potential contract that doesn't go to a competitor. Finally, the 10-year partnership horizon creates a predictable stream of infrastructure revenue regardless of how the AI model market itself develops.

For Anthropic, the terms of the agreement address a critical problem — the shortage of computational resources. Training frontier models requires enormous capacity, access to which is limited and expensive. The guaranteed 5 gigawatts removes one of the main operational risks for this fast-growing company and allows planning for training of next-generation Claude without worrying about GPU cluster availability.

At the same time, Anthropic formally retains its status as an independent lab — despite deep integration with the infrastructure and strategy of a single investor. What is happening reflects a key dynamic of the current AI race: competitive struggle unfolds not only at the level of model quality, but also over control of computational infrastructure. Cloud providers — Amazon, Microsoft, Google — are actively competing to make their platform the standard for training and deploying next-generation AI systems.

Amazon's investment in Anthropic is not simply a financial bet. It is an architectural decision about what the AI ecosystem will look like ten years from now: who owns the chips, who manages the cloud, who builds the models — and how these three levels will be connected to each other.