The social media giant is eyeing a new revenue stream by selling its massive AI compute capacity to outside customers, a move that could reshape cloud competition.
Mark Zuckerberg wants to rent out Meta’s AI brain. The Meta CEO confirmed during a July 9 interview with Bloomberg that the company is actively exploring an AI cloud business, calling it a logical next step given the sheer scale of infrastructure the company has already built.
Meta is on track to spend up to $145 billion on AI infrastructure in 2026 alone.
From social network to cloud contender #
Zuckerberg first floated the idea publicly at a May 27 shareholder meeting, saying the prospect of entering the cloud computing sector was “definitely on the table.” He noted that external companies inquire “almost every week” about purchasing compute or API access from Meta.
On July 1, Bloomberg reported that Meta was already building a new business unit specifically designed to monetize its excess AI computing capacity and AI models, including something called Muse Spark, for outside customers. The company is reportedly considering aggressive pricing models.
Wall Street liked what it heard. Meta shares jumped nearly 9% following the July 1 report on the cloud business plans.
A crowded but growing market #
If Meta pulls the trigger, it would be stepping into a ring with three entrenched heavyweights: Amazon Web Services, Microsoft Azure, and Google Cloud. Total Big Tech AI infrastructure investments are projected to exceed $700 billion in 2026. Microsoft has OpenAI as its AI ace. Google has Gemini and its own TPU chips. AWS has the largest market share and the deepest enterprise penetration.
What this means for investors and the broader market #
For Meta shareholders, the near-term signal is positive. The 9% stock jump after the initial report suggests the market views cloud revenue as a meaningful catalyst. But execution risk is real. Building an enterprise cloud business requires more than spare servers. It requires sales teams, SLAs, security certifications, and the kind of customer support infrastructure that Meta has never needed for its consumer products. Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our