Anthropic Signs $45 Billion Cloud Deal With Nscale for AI Compute

Anthropic, the AI safety company behind the Claude family of large language models, has struck a $45 billion cloud infrastructure deal with Nscale, a European data center and cloud provider, according to sources familiar with the matter. The agreement represents one of the largest compute procurement contracts in the rapidly accelerating race to secure AI processing power.

Under the terms of the deal, Anthropic will rent approximately 460 megawatts of compute capacity housed at an Nscale data center development in West Virginia. To put that figure in perspective, 460 megawatts is enough electricity to power hundreds of thousands of homes — a stark illustration of how energy-intensive frontier AI training and inference workloads have become.

Read more Calian Group Sells U.S. Commercial IT Unit to Trace3 →

The partnership signals a broader strategic shift in how leading AI labs are approaching infrastructure. Rather than relying solely on hyperscale cloud providers like Amazon Web Services or Microsoft Azure, companies such as Anthropic are increasingly turning to specialized compute providers to lock in dedicated capacity at scale, reducing dependency and potentially securing more favorable long-term economics.

For Nscale, landing Anthropic as an anchor tenant carries enormous commercial and reputational weight, potentially positioning the firm as a serious competitor in the high-demand AI infrastructure market. West Virginia, with its relatively low energy costs and available land, has been emerging as a quiet contender in the data center buildout sweeping the United States.

The sheer dollar magnitude of this arrangement — $45 billion — underscores how capital-intensive the AI industry has become and raises broader questions about concentration of compute resources among a handful of well-funded labs. Continue reading at US Top News and Analysis.