Meta follows SpaceX's playbook, looks to monetise excess AI compute
What's the story
Meta Platforms is planning to enter the cloud computing space by leasing out its excess artificial intelligence (AI) computing capacity, Bloomberg News has reported. The move could reduce Meta's reliance on advertising revenue and help it compete with major cloud players such as Amazon, Microsoft, and Alphabet. The company's shares rose nearly 6% in premarket trading after the news broke.
Strategic shift
SpaceX announced similar plans in May
One of the options being considered by Meta is to offer customers access to AI models on its existing infrastructure, much like Amazon Web Services (AWS) does with its Bedrock platform. This strategic shift comes after SpaceX, through xAI, announced similar plans. SpaceX signed a deal with Anthropic in early May to lease out the full compute capacity of its Colossus 1 data center. It has since inked similar agreements with Google and Reflection AI.
Future prospects
Zuckerberg hinted at possible cloud expansion in May
Earlier this year, Meta CEO Mark Zuckerberg had hinted at the company's possible entry into the cloud computing space if it overspends on data centers and has excess capacity. He said, "It's definitely on the table," during Meta's annual shareholder meeting in May. The tech giant is also said to be exploring a business model similar to CoreWeave's, selling access to raw compute capacity.
Infrastructure development
Meta's significant investment in AI infrastructure
Despite not seeing significant demand for its own AI models and services, Meta has invested heavily in infrastructure for AI compute. By the end of Q1 2026, the company had committed to spending $182.9 billion on AI infrastructure over the coming years. This includes massive ongoing projects in Louisiana and Ohio, with the latter expected to come online this year.
Project launch
Formation of Meta Compute and rising AI infrastructure spending
The new cloud business will be part of an initiative called Meta Compute, led by Santosh Janardhan, head of infrastructure at Meta, Daniel Gross, head of Superintelligence Labs at Meta, and president Dina Powell McCormick. The move comes as Big Tech firms are expected to spend over $700 billion on AI infrastructure this year, up from about $400 billion in 2025.