Compute Capacity Expansion
In a significant strategic maneuver, Anthropic has forged a substantial partnership with both Google and Broadcom to guarantee access to multi-gigawatt
quantities of next-generation Tensor Processing Units (TPUs). This critical infusion of computing power is slated to become available starting in 2027, a move designed to proactively address the burgeoning global demand for its sophisticated Claude AI models. This agreement represents one of the most significant infrastructure investments made by Anthropic to date, underscoring the company's rapid ascent and its commitment to staying ahead in the fiercely competitive artificial intelligence landscape. The enhanced capacity will not only support Anthropic's future pioneering models but also significantly improve the performance and reach of its services for users worldwide, ensuring continued innovation and service excellence as demand escalates.
Financial Growth Trajectory
This strategic infrastructure expansion is happening in parallel with Anthropic's impressive financial performance. The company has announced that its annualized revenue run rate has surpassed an astonishing $30 billion in 2026, a remarkable jump from approximately $9 billion at the close of 2025. Furthermore, Anthropic revealed it now serves over 1,000 enterprise clients who each spend upwards of $1 million annually, a client base that has impressively doubled in just two months. This robust financial growth is a testament to the increasing adoption and value derived from Anthropic's AI solutions, particularly its Claude models, and provides a strong foundation for continued investment in cutting-edge technology and infrastructure.
Strengthening Domestic Infrastructure
A substantial portion of the newly acquired computing power will be strategically located within the United States, aligning with Anthropic's previous commitment announced in November 2025 to invest $50 billion in domestic computing infrastructure. This focus on U.S.-based resources is a deliberate step towards building more resilient and secure AI systems. The collaboration with Google further solidifies Anthropic's existing relationship with Google Cloud and builds upon earlier agreements for TPU capacity. Simultaneously, the ongoing partnership with Broadcom continues to drive innovation at the hardware level, ensuring that Anthropic has access to state-of-the-art components necessary for its advanced AI development and deployment.
Multi-Cloud Strategy Focus
Despite these significant partnerships, Anthropic remains committed to a multi-cloud strategy, utilizing a diverse array of AI hardware platforms. This includes leveraging systems from Amazon Web Services, Google's TPUs, and NVIDIA GPUs. This flexible approach allows the company to optimize workload distribution, enhance application reliability, and maintain agility in a rapidly evolving technological environment. Anthropic has affirmed Amazon as a core cloud and training partner, with ongoing work on Project Rainier. This multi-platform presence positions Anthropic as one of the few AI firms with Claude models available across all three major cloud providers: AWS, Google Cloud, and Microsoft Azure.
