Anthropic signs a major deal with Google Cloud: Google demonstrates its strength, Amazon faces pressure

```

Author of this article: Long Yue

Source: Hard AI

The battle for computing power in the AI field is intensifying, and Google has achieved a major victory in this key contest. Its cloud business department has won a big order from AI unicorn Anthropic, which will not only bring considerable revenue growth to Google Cloud, but also put direct pressure on its main competitor, Amazon.

According to the latest report from Morgan Stanley, Anthropic has officially announced the expansion of its collaboration with Google Cloud. This “milestone” agreement stipulates that Anthropic will receive access to as many as 1 million Google TPU chips to support the training and service of its next-generation Claude models.

Morgan Stanley’s analysis points out that the total value of this deal is as high as “tens of billions of dollars.” The report further estimates, based on similar deals such as Google’s contract with Meta, that the term of this agreement may be six years, with a total value between $50 billion to $80 billion.

Anthropic is expected to have more than 1 gigawatt (GW) of online computing power by 2026. The report believes that the strong cost performance and cost efficiency of Google’s TPU is the key factor attracting Anthropic, especially as Anthropic’s own business is expected to achieve about 150% compound annual growth rate between 2025 and 2027.

A Shot in the Arm for Google Cloud: Annual Revenue in the Billions, Accelerated Growth

For Google, this deal is a major validation of its AI cloud strategy. Morgan Stanley predicts that this will become a key driver for accelerating Google Cloud Platform (GCP) revenue growth in 2026 and beyond.

The report’s mathematical model shows that this cooperation is expected to bring an additional 100 to 900 basis points increase to Google Cloud’s revenue growth in 2026. More importantly, by 2027, when the partnership is in full swing, it is expected to contribute a stable about $9 to $13 billion in annual revenue to Google Cloud.

Analysts Brian Nowak and Joseph Moore believe the key factors in Anthropic’s selection of Google TPUs are their strong value for money and cost effectiveness, which is crucial for Anthropic at this stage of rapid expansion.

Pressure on Amazon: Main Partner Status at Stake

For a long time, AWS has been Anthropic’s most important infrastructure support. However, Google Cloud’s involvement this time breaks AWS’s exclusive advantage.

Analysts compare the approximately 1GW of computing power offered by Google to the approximately 2GW of capacity AWS provided for Anthropic at the “Project Rainier” data center in Indiana.

This comparison indicates that while AWS still holds about two-thirds of the market share advantage, its failure to secure this key incremental order has raised questions about its technological competitiveness, capacity, or pricing strategy.

The report points out that “AWS needs to continue to prove its computing capacity and efficiency (including performance and price)” to cope with the increasingly fierce competition.

Inference vs. Training: Google TPU’s Differentiated Advantage

The report also reveals an important technical detail: the computing power provided by Google Cloud this time will be mainly used for “inference” workloads, rather than “training.”

Morgan Stanley’s analysis gives two reasons: First, Anthropic has made it clear that AWS is its main “training” partner; Second, Google’s TPU v7 (codename Ironwood), which is expected to be widely deployed in 2026, was designed specifically for efficient inference tasks.

This shows that Google is establishing a strong moat in specific segments of the AI workflow with its custom AI chips, thereby opening up a differentiated competitive path in a market dominated by Nvidia GPUs.

This article is from public WeChat account “Hard AI”. For more cutting-edge AI news, please go here

Risk Warning and DisclaimerThe market has risks, investment needs caution. This article does not constitute personal investment advice, nor does it take into account the special investment objectives, financial status, or needs of individual users. Users should consider whether any opinions, viewpoints, or conclusions in this article are suitable for their particular situation. Investment based on this is at your own risk. ```