Microsoft, Nvidia and Anthropic ink $45 billion deal that furthers a big critique of the AI trade

By Emily Bary

Anthropic will buy services from Microsoft and Nvidia, which are investing in the company. That could stoke additional concerns about circular financing among AI players.

Anthropic operates the Claude large language model.

Another week, another megadeal involving Big Tech players jockeying to power the artificial-intelligence wave - with large sums of money moving between them.

The new arrangement between Microsoft (MSFT), Nvidia (NVDA) and Anthropic has two components. Anthropic, which is the operator of the Claude AI large language model, has agreed to buy $30 billion worth of compute capacity from Microsoft's Azure cloud business. Meanwhile, Nvidia plans to invest up to $10 billion in Anthropic, while Microsoft will kick in its own investment of up to $5 billion.

That aspect could further concerns about circular financing within the AI ecosystem, referring to the idea that companies are investing in AI players that will then pay them for services.

Don't miss: Big Tech needs a staggering $1.5 trillion to fund the AI boom. This is the complex playbook it's using to get it.

The partnership also highlights how AI upstarts like Anthropic are diversifying to ensure they have as much cloud capacity and compute power at a time when both are in short supply. Amazon.com (AMZN) and Alphabet (GOOGL) (GOOG), which operate rival cloud businesses to Microsoft, are already investors in Anthropic and provide services to the company.

With the arrangement, Claude will be "the only frontier LLM model available on all three of the world's most prominent cloud services," according to a press release.

The companies say the partnership will mean more model options for Azure customers. Additionally, the Nvidia component is meant to give Anthropic's models "the best possible performance, efficiency and TCO," or total cost of ownership. Anthropic is committing to up to 1 gigawatt of Nvidia compute capacity at the outset.

The growing interconnectedness of AI players has stoked concerns about a possible bubble. Megadeals with companies like OpenAI and Anthropic can mean big money for cloud-computing and hardware players, but they also add risk if the suppliers are rapidly building out capacity to support startups that still have much work to do in terms of making AI services work economically.

See more: Has AI gotten too big to fail? How the U.S. government is backstopping the tech boom.

-Emily Bary

This content was created by MarketWatch, which is operated by Dow Jones & Co. MarketWatch is published independently from Dow Jones Newswires and The Wall Street Journal.

(END) Dow Jones Newswires

11-18-25 1142ET