AI research powerhouse Anthropic has announced a significant expansion of its computing capacity through new agreements with Google and Broadcom. This move is designed to meet the rapidly accelerating demand for its Claude AI models, which have seen explosive growth across both consumer and enterprise sectors.
Expanding the Compute Footprint
The new deal builds upon a previous agreement established in late 2025, which secured over one gigawatt of compute capacity. Under the updated terms, Anthropic will expand its use of Google Cloud’s Tensor Processing Units (TPUs) —specialized hardware designed specifically to accelerate machine learning workloads.
While Anthropic has not released specific technical details, recent SEC filings from Broadcom suggest the scale of this expansion is massive, potentially encompassing 3.5 gigawatts of compute. This infrastructure is slated to come online in 2027.
A Strategic Investment in U.S. Infrastructure
A central component of this expansion is a commitment to domestic growth. Anthropic stated that the majority of this new compute capacity will be housed within the United States. This aligns with the company’s broader $50 billion pledge to invest in U.S.-based computing infrastructure.
This trend highlights a critical shift in the AI industry: as models become more sophisticated, the bottleneck for growth is no longer just software, but the physical availability of energy and hardware. By securing long-term access to massive power supplies and specialized chips, Anthropic is attempting to “future-proof” its ability to train and deploy next-generation models.
Explosive Financial and User Growth
The massive infrastructure push follows a period of unprecedented financial scaling for the company:
- Revenue Surge: Anthropic’s run rate revenue has climbed to $30 billion, a dramatic increase from the $9 billion recorded at the end of 2025.
- Enterprise Adoption: The company now boasts over 1,000 business customers who spend more than $1 million annually on its services.
- Valuation Leap: Following a recent $30 billion Series G funding round, Anthropic’s valuation has reached $380 billion.
Despite facing regulatory scrutiny—including being labeled a supply-chain risk by the U.S. Defense Department—the company’s market momentum remains strong.
“We are building the capacity necessary to serve the exponential growth we have seen in our customer base while also enabling Claude to define the frontier of AI development,” said Krishna Rao, CFO of Anthropic.
Why This Matters
This deal signals that the “AI arms race” has moved into a heavy industrial phase. For Anthropic to remain competitive against rivals like OpenAI, it cannot rely solely on algorithmic breakthroughs; it must secure the massive amounts of electricity and specialized silicon required to run them. The partnership with Google and Broadcom ensures that Anthropic has the hardware foundation to support its rapidly growing enterprise client base through the end of the decade.
In summary, Anthropic is aggressively securing the massive computational and energy resources needed to sustain its rapid commercial expansion and maintain its position at the forefront of AI development.
