Mark Zuckerberg-led Meta Platforms is reportedly building a cloud business to sell its excess artificial intelligence (AI) computing capacity, news that comes amid the global tech giants seeking returns on costly AI investments amid overspending worries.
“The plans are still in development, and the strategy could change,” a Bloomberg report said, citing sources familiar with the matter.
At Meta’s shareholder meeting in May, Zuckerberg had said entering cloud computing was “definitely on the table” for the social media conglomerate, noting that firms were approaching Meta “almost every week” to buy access to its AI models or spare computing power.
While competing against top cloud providers Amazon, Microsoft, and Alphabet may help Meta tap booming demand for AI services from businesses, its new business plan, if it proceeds, will also reduce the tech giant’s reliance on the advertising market.
However, analysts doubt Meta’s capability to catch up with leading AI labs such as Anthropic, despite Zuckerberg backing the tech giant’s AI computing ambitions with billions of dollars in investment, including setting off a talent war in 2025.
Meta, in April, unveiled “Muse Spark,” the first AI model from a costly team it assembled but one it is yet to release to developers. A Wall Street Journal (WSJ) report said in June that the tech biggie has no scheduled date for the launch.
“The cloud service Meta is now planning would let developers access AI models hosted on its infrastructure, including Muse Spark, and pay for the computing power needed to run them. This would be similar to Amazon Web Services’ Bedrock, which allows developers to access AI models from different companies,” Bloomberg reported.
“The impact of adding Meta’s capacity to the market is more likely to be on neoclouds than the big hyperscalers. Those companies like CoreWeave and Nebius rely on Meta for their growth, and Meta may not need them anymore,” said Gil Luria, managing director of investment banking company DA Davidson.
“This is very similar to the situation SpaceX has found itself in, which led it to sell compute capacity as well,” he said further.
Elon Musk’s SpaceX, whose xAI unit is also behind leading AI development efforts, recently struck deals to rent out access to its Memphis data centers to Anthropic and Google.
As per the reports, Meta is projected to spend as much as USD 145 billion on AI infrastructure in 2026, a significant portion of Big Tech’s more than USD 700 billion outlay on the technology.