Anthropic invests 50 billion dollars in American AI infrastructure
Anthropic has announced a massive investment plan of 50 billion dollars to strengthen American computing infrastructure.
This initiative, in partnership with the cloud platform Fluidstack, aims to build several data centers across the country, with confirmed locations in Texas and New York — and “more to come.”
Anthropic: Data Centers to Support Claude’s Growth
According to Anthropic, the new centers will gradually go online starting in 2026 and are expected to create around 800 direct jobs.
The objective is twofold: to meet the growing demand for its AI assistant Claude and to support cutting-edge research in artificial intelligence.
“This expansion is crucial for keeping our research at the forefront of what’s possible and ensuring Claude’s long-term performance and safety,” the company stated in its announcement.
A Strategic Investment Aligned with American Tech Policy
Anthropic claims that this project is part of the AI Action Plan initiated by the Trump administration, which aims to maintain the technological leadership of the United States in artificial intelligence and bolster domestic capabilities in this critical sector.
These new data centers will be built based on energy-efficient and modular designs, capable of hosting the most advanced AI architectures.
An Unprecedented Race for AI Infrastructure
Anthropic joins the ranks of tech giants making massive bets on AI infrastructure:
- OpenAI and SoftBank announced their “Stargate” project in January, valued at an estimated 500 billion dollars, with an initial site in Texas.
- Meta, meanwhile, has already committed 600 billion dollars to developing infrastructure and data centers in the United States.
This race for computing power illustrates the growing dependence of generative AI models on vast and specialized hardware resources.
A Response to Explosive Demand for Computing Power
Anthropic emphasizes that this investment “is necessary to meet the growing demand” generated by Claude — its language model that competes with ChatGPT — used by millions of businesses and individuals worldwide.
The new data centers will enhance computing capacity for training and inference of models, reduce latency for users, and ensure better technological sovereignty for its American clients.
Looking Ahead
This announcement confirms that the new battle in AI is focused on infrastructure. As models become more powerful and costly to operate, mastery of high-performance computing centers becomes a national strategic asset.
With this 50 billion dollar plan, Anthropic positions itself not only as a major player in AI but also as a pillar of American digital sovereignty in the face of global competition.
