Anthropic plans to invest $50 billion to build data centers dedicated to artificial intelligence work in multiple locations in the U.S., including Texas and New York, as part of its latest $1 billion infrastructure effort to support the AI boom.
- Activities in the United States: Google accused of using Gemini AI tool to spy on users
- Administrator and password: See the ranking of the most common passwords of 2025 and how they facilitate cybercrime.
The new center, which Anthropic is developing in partnership with UK-based Fluidstack, is expected to begin operations throughout 2026, the company announced in a statement Wednesday.
The project will be the first major data center expansion led directly by Anthropic, rather than through a cloud computing partner such as Amazon or Alphabet Inc.’s Google.
Rival OpenAI, the developer of ChatGPT, said the data center project advances the Trump administration’s goal of “maintaining U.S. leadership in AI” by strengthening the country’s technology infrastructure.
Anthropic said it aims to create 800 permanent jobs and 2,400 construction jobs at the new site. Fluidstack, a startup that sells cloud computing services for AI, will provide “gigawatts of power” as part of the project, according to a statement.
— We are moving ever closer to an AI that can accelerate scientific discovery and solve complex problems in ways not previously possible — said Dario Amodei, CEO and co-founder of Anthropic. — These locations will help develop more advanced AI systems, accelerate these discoveries, and create American jobs.
Billions of dollars in donations and investor doubts
Companies like OpenAI, Meta (owner of Instagram and Facebook), Google, Microsoft, Nvidia, and other tech giants plan to jointly invest hundreds of billions of dollars in data centers to support AI.
Earlier this year, OpenAI announced a $500 billion investment in the U.S. project under the Stargate brand, which has since expanded to other countries. Meta is building a 2GW facility in rural Louisiana, and CEO Mark Zuckerberg recently said the company plans to invest $600 billion in data centers, infrastructure and jobs in the U.S. over the next few years.
- For Europe: Google announces US$6.4 billion investment in Germany
These investments come amid investor doubts that the spending is excessive for a technology that has yet to demonstrate a profitable business model. Sarah Friar, OpenAI’s chief financial officer, downplayed the concerns last week, saying, “I don’t think there’s enough enthusiasm for AI when you think about the practical implications and what AI can do for humans.”
Founded in 2021 by former OpenAI employees, Anthropic positions itself as a trusted, security-focused company. Although smaller than its rivals, Claude chatbot and its underlying technology are gaining traction among developers as well as enterprise customers in finance and healthcare.
In September, the company raised $13 billion, valued it at $183 billion, and claims to have 300,000 enterprise customers.
- New project: Microsoft invests US$10 billion in AI data center in Portugal
Fluidstack, Anthropic’s partner in data center expansion, is one of the so-called “neoclouds,” or companies that rent computing power for AI development. Earlier this year, Fluidstack inked two unusual deals with Google involving crypto miners (TeraWulf and Cipher Mining). In both cases, Google acted as a guarantor, promising to repay creditors if Fluidstack defaulted or declared bankruptcy.
Fluidstack also played a central role in French President Emmanuel Macron’s AI initiative. In February, the company announced it would build a 1 gigawatt supercomputer in France at a cost of 10 billion euros ($11.5 billion), scheduled to be operational in 2026.
About The Author
