AI startup Anthropic has reportedly committed to spending approximately $200 billion over five years on Google Cloud infrastructure and AI chips, marking one of the largest cloud computing agreements ever signed in the artificial intelligence industry.
According to reports, the agreement represents more than 40% of the cloud revenue backlog recently disclosed by Alphabet, highlighting the growing importance of AI companies in driving hyperscale cloud demand. The backlog reflects long-term contractual commitments from enterprise cloud customers.
As part of the deal, Anthropic is expected to receive access to multiple gigawatts of AI computing capacity powered by Google’s tensor processing units (TPUs), developed in collaboration with chip partner Broadcom. Reports indicate this infrastructure will begin coming online from 2027 onward.
The agreement further deepens the relationship between Google and Anthropic, despite the two companies also competing in the AI market. Alphabet is reportedly investing up to $40 billion in Anthropic as part of broader strategic partnerships centered on AI infrastructure, cloud services, and model deployment.
Anthropic has been aggressively expanding its computing footprint to support surging demand for its Claude family of AI models. In recent months, the company has signed multiple infrastructure agreements across the industry, including partnerships with Amazon Web Services, CoreWeave, and Nvidia-powered cloud providers. Anthropic currently trains and operates its models using a mix of Google TPUs, AWS Trainium chips, and Nvidia GPUs.
The scale of the commitment underscores the unprecedented surge in AI infrastructure spending across the technology industry. Analysts estimate that major technology firms could collectively spend more than $700 billion on AI infrastructure in 2026 alone as competition intensifies around large language models, AI agents, and enterprise AI services.
The deal also reinforces Google Cloud’s growing position in the AI market. Alphabet recently reported record cloud growth driven largely by enterprise AI adoption and increasing demand for its custom AI chips and infrastructure services.
