Anthropic signed one of the largest artificial intelligence infrastructure contracts in history: a 20-year deal with TeraWulf for a 401 MW data center in Kentucky, which would be operational by 2028. It's estimated that TeraWulf will generate around USD 19 billion in revenue from the deal. Here we explain what this move means for the race for AI infrastructure and how it affects the competition among the big players.

The announcement, reported on July 7 by media like The Verge and The Wall Street Journal, is not just a financial news item: it's a signal of where the industry is heading. AI is no longer played only in the models, but in who has the energy and the data centers to train and operate them.

What the deal says

• Duration: 20 years, an enormous commitment that reflects the magnitude of the computing needs.

• Capacity: a 401 MW data center, a scale reserved for the largest projects in the sector.

• Location and timeline: Kentucky, with operation estimated by 2028.

• Economic magnitude: some USD 19 billion in projected revenue for TeraWulf.

Why infrastructure is the new battlefield

Training and operating frontier AI models requires an amount of computing and energy that was unthinkable a few years ago. The competitive advantage is no longer only in having the best model, but in securing stable access to energy and data center capacity over the long term. That's why AI companies sign giant multi-year contracts: they are securing the fuel for the next decade. Anthropic, with this move, rises to the level of the infrastructure that OpenAI and Google deploy.

How it affects the competition

The deal repositions Anthropic against its rivals. While OpenAI (backed by Microsoft) and Google (with its own infrastructure) have access to massive computing, Anthropic gives a clear signal that it will compete in the same league. For the ecosystem, this means more available capacity for increasingly powerful models, but it also concentrates power in the few players capable of signing contracts of this scale. It's a trend we've been following in the news about Anthropic and its Claude ecosystem and in the analysis of the geopolitics of AI.

To follow the detail, media like The Verge cover the race for AI infrastructure, and consulting firms like McKinsey analyze how investment in computing conditions the pace of AI adoption in companies. The underlying message: infrastructure defines who can innovate.

What it means for professionals

More infrastructure means more powerful and accessible models, which accelerates the demand for profiles who know how to apply them. The race for data centers, although it may seem distant, ends up impacting the tools you use and the jobs created around AI. Understanding these trends helps you anticipate where it's a good idea to train.

Recommended Coderhouse courses

If this race for infrastructure shows you that AI is here to stay, it's a good time to train:

• Introduction to Artificial Intelligence Course: to understand the complete landscape of AI and its players.

• AI Engineering Course: if you want to build solutions on the models that this infrastructure makes possible.

• DevOps & Cloud Course: to understand the infrastructure and deployment layer behind AI.

Get ahead of the trend: train in AI and in the infrastructure technologies that are defining the next decade of the industry.

Frequently asked questions

What did Anthropic sign with TeraWulf?

A 20-year deal for a 401 MW data center in Kentucky, operational by 2028, with projected revenue of some USD 19 billion for TeraWulf. It's one of the largest AI infrastructure contracts in history.

Why is infrastructure so important for AI?

Because training and operating frontier models requires enormous amounts of computing and energy. Securing data centers and energy over the long term became a competitive advantage as key as the model itself.

How does this affect the competition among AI companies?

It repositions Anthropic against OpenAI and Google, showing that it will compete with infrastructure of the same scale. It also concentrates power in the few players capable of signing deals like this.

What does it imply for those who work or study technology?

More infrastructure accelerates more powerful and accessible models, which increases the demand for profiles who know how to apply AI. Anticipating these trends helps to decide what to train in.