Meta prepares a cloud business to sell its excess AI compute capacity

The Bloomberg article, written by Riley Griffin and Kurt Wagner, reveals that Meta Platforms is developing plans to create a cloud infrastructure business that would sell third parties access to AI computing capacity and to its own models. According to sources familiar with the matter who asked not to be identified because the details are not yet public, the company is thus seeking to open a new front of direct competition with the sector's established giants: Amazon Web Services, Microsoft Azure and Google Cloud.

The logic behind this move is fairly straightforward: Meta has been investing colossal sums in data centers and other infrastructure to sustain its own artificial intelligence ambitions—the article mentions in passing a $200 billion bet on a data center in Louisiana as an example of the scale of these investments. Given that level of spending, it makes financial sense for the company to seek to monetize its surplus computing capacity by reselling it to external customers rather than leaving it idle.

This move is significant because it marks a shift in stance for Meta, which has traditionally been a consumer of cloud services (even from its own rivals) rather than a provider. If confirmed and developed, the company would go from being just another AI infrastructure customer to becoming a player that competes directly for that market, leveraging the massive investments it is already making in chips, data centers and energy capacity for its generative AI products and its advertising operations.

It is important to note that the available content of this article is limited: the original Bloomberg piece is behind a paywall, and the captured text mostly includes site navigation elements, institutional links and menus, without developing much beyond the opening paragraphs cited above. No additional details are available on concrete timelines, the pricing model, whether it will launch as a standalone brand, or official statements from Meta on the matter, so any further speculation about scope, timeline or competitive impact would exceed what the material allows to be stated with confidence.