Anthropic has announced a significant investment of $50 billion to construct artificial intelligence infrastructure within the United States. The project will commence with the development of custom data centres in Texas and New York, designed to support the company’s rapid enterprise growth and long-term research objectives. These facilities will be built in partnership with Fluidstack, an AI cloud platform providing large-scale graphics processing unit (GPU) clusters to clients such as Meta, Midjourney, and Mistral.
Additional sites are planned, with the first locations slated to go live in 2026. The project is expected to generate 800 permanent jobs and over 2,000 construction roles. Anthropic’s investment positions it as a major domestic player in physical AI infrastructure, aligning with policymakers’ increasing focus on U.S.-based computing capacity and technological sovereignty. CEO Dario Amodei stated that this infrastructure is crucial for developing more capable AI systems that can drive scientific breakthroughs.
Anthropic serves over 300,000 businesses, with enterprise clients contributing the majority of its revenue. The company specialises in AI safety and research, working to build reliable, interpretable, and steerable AI systems. The number of large accounts, generating more than $100,000 annually, has grown nearly sevenfold in the past year. In parallel, Amazon has opened a dedicated data centre campus for Anthropic on 1,200 acres in Indiana. This $11 billion facility is already operational.
The investment occurs as the role of the federal government in financing AI infrastructure becomes a focal point. Recently, OpenAI requested the Trump administration to broaden a key CHIPS Act tax credit to encompass AI data centres and grid components, highlighting the financial and political uncertainties surrounding the funding of America’s AI infrastructure.
