Meta Nears $10B AI Compute Deal With Anthropic
Meta is reportedly in advanced negotiations to lease AI computing capacity from its data centers to Anthropic in a deal valued at around $10 billion. If completed, the agreement would become one of the first major commercial customers for the company’s emerging cloud infrastructure business, internally known as Meta Compute.
The potential partnership follows recent efforts to expand beyond internal AI development. Instead, the company aims to generate revenue by offering AI computing resources and, eventually, AI models to external organizations.
Expanding Into AI Infrastructure Services
The cloud infrastructure initiative is led by head of infrastructure Santosh Janardhan, Meta Superintelligence Labs leader Daniel Gross, and president Dina Powell McCormick. Earlier this year, the company indicated that selling compute capacity to external customers remained part of its long-term strategy.
Meanwhile, the planned business would compete with major cloud providers while also exploring the sale of AI models hosted on its own infrastructure. As a result, the strategy could diversify the company’s AI revenue beyond consumer products and advertising.
Competition Heats Up in the AI Compute Market
The reported negotiations reflect a broader industry trend toward large-scale AI compute leasing. Similar long-term agreements have already emerged as AI companies seek reliable access to high-performance infrastructure for model training and deployment.
At the same time, the company continues to invest heavily in expanding its AI infrastructure. It plans to build tens of gigawatts of data center capacity over the coming years, while capital expenditure for 2025 is projected between $64 billion and $72 billion. Additionally, it recently unveiled a $10 billion hyperscale data center in Alberta, Canada, its largest facility outside the United States.
What’s Next?
Although negotiations are reportedly in their final stages, the agreement has not been finalized. Therefore, both the deal and the broader cloud infrastructure strategy could still change before an official announcement.
Why Does It Matter?
A successful agreement would mark a significant shift from building AI infrastructure solely for internal use to offering commercial AI computing services. Moreover, it would position Meta as a direct competitor in the rapidly expanding AI cloud market, where demand for high-performance compute continues to surge.
Looking Ahead
If finalized, the partnership could accelerate the commercialization of Meta Compute and strengthen competition across the AI infrastructure industry. As enterprises seek greater access to advanced computing power, large technology companies are increasingly transforming their massive AI investments into new revenue-generating cloud businesses.
