Meta Challenges AWS and Google Cloud in Cloud Computing

Mark Zuckerberg, CEO of Meta. Image credit: Getty Images

Mark Zuckerberg confirms that selling computing capacity is an option as Meta Compute prepares to enter the cloud market, competing with Amazon and Google.

Meta has announced plans to launch a cloud infrastructure business, positioning the tech company to compete against established services like Amazon Web Services, Microsoft Azure, and Google Cloud. The initiative aims to give external customers direct access to AI models and computing power.

An internal group called Meta Compute will play a key role in these efforts, formed to oversee the development and management of Meta's AI infrastructure.

Meta Compute is led by several company executives, including Santosh Janardhan, Meta’s Head of Infrastructure; Daniel Goss, an executive within Meta’s Superintelligence Labs AI Unit; and Meta President Dina Powell McCormick.

One potential model under consideration for the computing segment would allow outside developers to pay for running queries against AI models on infrastructure owned and operated by Meta. These models include Meta’s proprietary generative AI model, named Muse Spark.

In addition, Meta plans to create a separate channel to rent out raw GPU capacity directly to its customers.

KEY FIGURES

• Meta projects spending up to US$145 billion on AI infrastructure this year
• The technology industry averages US$700 billion in AI technology spending

Selling computing access to external developers

Speaking to shareholders in May, Meta CEO Mark Zuckerberg said the company has been considering the idea of selling computing access. He added that the concept is “definitely on the table.”

“Almost every week there are different companies that come to us from the outside asking us to both stand up an API service or asking if we have compute that they could buy from us at some premium to what we've bought it at,” Mark said.

He explained that no external deals have been made so far because internal demand has consumed all available capacity. However, the CEO noted that if overbuilding occurs, Meta will move to sell the extra capacity externally.

Mark Zuckerberg, CEO of Meta, says selling computing access to other companies is “definitely on the table.” Credit: Getty

Investing into AI superintelligence infrastructure

Superintelligence has become a major priority for Meta as it invests hundreds of billions of dollars into AI infrastructure. Meta has also secured large capacity agreements with CoreWeave, Google, and Oracle.

Some investors have questioned how such operations will generate revenue. Despite these doubts, the potential cloud business could prove profitable for the organization.

Compute supply constraints, a key point of discussion, first came into focus when Google restricted access to its Gemini AI. Google stated it could not meet Meta’s demand for AI compute.

This supply limitation delayed some internal AI efforts at Meta, leading the company to ask its own employees to reduce their AI token consumption.

Meta AI, the company’s official AI assistant. Credit: Meta

Muse Spark has since taken over a large portion of the work originally handled by Gemini. This shift comes as Meta begins to develop and expand its own in-house AI capabilities.

The company first unveiled the model in April, but it has not yet been released to developers. A confirmed release date for the AI model is still pending, The Wall Street Journal reports.

Meta is projected to spend up to US$145 billion on AI infrastructure this year. This expenditure represents a significant share of the average US$700 billion spent on AI technology across the tech industry.