Microsoft and Nvidia forge multi-billion dollar alliance with Anthropic

• Anthropic will use Azure infrastructure and Microsoft will use their models to expand go-to-market initiatives together.
• Anthropic and Nvidia will collaborate on specialised AI chips and models.

Microsoft and Nvidia announced a substantial new investment collaboration with Anthropic, the AI start-up behind Claude, as part of a deal that will see Anthropic commit a landmark $30 billion to Microsoft’s Azure cloud services.

The partnership deepens ties among some of the most influential players in artificial intelligence, reflecting the industry’s skyrocketing demand for computing power and strategic alliances.

According to company statements, Nvidia is set to invest up to $10 billion in Anthropic, while Microsoft’s contribution could total as much as $5 billion. While specific terms remain under wraps, sources indicate these commitments are tied to Anthropic’s upcoming funding round.

“This partnership signals a pivotal shift in the AI ecosystem,” Microsoft CEO Satya Nadella stated in a video address.

“We’re increasingly going to be customers of each other—Anthropic will use our infrastructure, and we will use their models as we expand go-to-market initiatives together.” Nadella emphasised Microsoft’s ongoing commitment to OpenAI, referring to it as a “critical partner.”

A formidable contender

The announcement comes on the heels of sweeping changes at OpenAI and a significant $38 billion cloud services agreement between the ChatGPT maker and Amazon.com, as AI market leaders seek enhanced operational autonomy and scale. OpenAI recently disclosed plans to invest $1.4 trillion into expanding its compute capabilities, aiming for 30 gigawatts in infrastructure.

With its latest moves, Anthropic—founded in 2021 by former OpenAI employees and now valued at $183 billion—continues to position itself as a formidable contender.

The company anticipates its revenue run rate could soar to $28 billion—next year, with over 300,000 business clients already on board.

As part of the multi-faceted deal, Anthropic and Nvidia will collaborate on specialised AI chips and models. Anthropic has agreed to utilise up to 1 gigawatt of Nvidia’s high-performance hardware, a capacity valued in the range of $20–$25 billion according to industry experts.

Notably, the deal will allow Microsoft’s Azure AI Foundry clients access to Claude, Anthropic’s impressive AI model, making it the only “frontier” AI model available across the world’s three largest cloud platforms. Despite these moves, Amazon will remain Anthropic’s primary cloud provider and training partner.

Will the AI bubble burst?

Many industry predict that the AI bubble is likely to burst and at last week’s Cerebral Valley Summit in San Francisco, more than 300 AI founders and investors named Perplexity as the billion-dollar startup most likely to fail, with industry heavyweight OpenAI coming in a close second.

Perhaps most notable at the Summit was the shifting investor sentiment toward Anthropic, which attendees named the top pick for future investment over OpenAI.

Despite a consensus that OpenAI will likely lead next year’s global LMArena leaderboard for advanced AI models, Anthropic’s growth has stolen the spotlight.

Moreover, OpenAI’s conversion challenges remain acute: 95 per cent of ChatGPT’s 800 million users use the service for free. Unlike Anthropic, which aims to break even by 2028, OpenAI is not expected to do so until the decade’s end.

In an interview with the BBC published on Tuesday, Pichai described the ongoing AI investment wave as an “extraordinary moment” but acknowledged “elements of irrationality” reminiscent of the 1990s tech bubble. “I think no company is going to be immune, including us,” Pichai said, when pressed on how Google might weather an industry downturn.

Analysts have increasingly debated whether current AI valuations are sustainable, and Pichai’s comments add momentum to a growing chorus warning of “irrational exuberance” in the sector.