Social media giant Meta Platforms (META) is brewing a major business transformation. Following reports that Alphabet (GOOGL) subsidiary Google has restricted its AI computing access, market sources indicate that Meta is planning to enter the fiercely competitive cloud infrastructure market. The company intends to package the excess AI computing power and model access accumulated from hundreds of billions of dollars in investments into new cloud service products for external sale, directly competing with Amazon’s AWS, Microsoft Azure, and Google Cloud for market share.<\/p>\n
This initiative not only signals Meta’s potential shift from a pure social media and advertising company to a cloud service provider but also represents the most concrete commercialization answer amid intensifying investor scrutiny over its AI capital expenditures. Following the news, Meta’s shares surged more than 8% in Wednesday’s pre-market trading before paring gains to around 6%. Meanwhile, concerns over heightened competition sent shares of emerging cloud service providers such as Nebius (NBIS.US), CoreWeave (CRWV.US), and IREN Ltd (IREN.US) tumbling over 6%, 5%, and 4% respectively in pre-market trading.<\/p>\n
Dual-Track Approach: Model Hosting and Bare Compute Leasing<\/p>\n
According to people familiar with the matter, Meta is currently coordinating the development of this cloud business through an internal initiative called “Meta Compute.” The core objective of the project is to extend the work of building and managing the company’s overall AI infrastructure to the external export of excess computing power.<\/p>\n
In exploring business models, Meta is primarily considering two potential directions. The first involves selling access to various AI models to external developers, operating similarly to AWS’s Bedrock service. Under this framework, Meta would be responsible for operating the data centers and underlying chips powering these models and would charge developers for access, including to its self-developed Muse Spark model. The second direction involves directly selling “raw” computing power to external customers, a model similar to the business logic of emerging cloud service providers like CoreWeave, allowing customers to directly lease Meta’s underutilized hardware infrastructure.<\/p>\n
The project is understood to be jointly led by a cross-departmental team of senior executives, with core members including Meta’s head of infrastructure Santosh Janardhan, senior leader from Meta’s super-intelligence lab AI division Daniel Gross, and Meta President Dina Powell McCormick. This underscores the high level of importance Meta places on this business transformation.<\/p>\n
Google’s Compute Curbs Highlight Urgency to Reduce Dependence<\/p>\n
The news of Meta’s cloud market entry comes amid very real pressures. Earlier this week, reports emerged that Alphabet had imposed access restrictions on Meta’s use of its Gemini AI technology. According to foreign media reports, because Google’s own computing capacity was insufficient to meet demand, it imposed usage caps on multiple customers, including Meta.<\/p>\n
This restriction has affected several of Meta’s internal operations, including areas such as content moderation and fraud detection, which previously relied heavily on Google’s AI technology. Following the access curbs, Meta has instructed employees to be more cautious when using AI features and to accelerate the migration of workloads to its internally developed Muse Spark model to reduce dependence on external AI service providers.<\/p>\n
Wedbush Securities analyst Matt Bryson commented on the situation, stating that this incident once again demonstrates that demand for AI computing continues to outstrip available supply. Despite technology companies investing massive capital expenditures to expand AI infrastructure, investment levels remain insufficient to keep pace with demand growth. Bryson specifically highlighted the danger of relying on companies that are simultaneously competitors to allocate resources. He noted that companies like Anthropic and Meta, which use Google’s cloud platform or its proprietary chips (TPUs), may face similar challenges in the future.<\/p>\n
Easing AI Investment Pressure, Finding an Outlet for Massive Spending<\/p>\n
Meta’s foray into cloud computing is essentially aimed at alleviating the enormous cost pressures brought by its aggressive AI expansion strategy. Meta CEO Mark Zuckerberg has established developing AI “super-intelligence” as the company’s top priority. To this end, the company has been racing to lock down expensive data centers and other necessary infrastructure resources, even reaching significant computing resource agreements with companies like CoreWeave, Google, and Oracle.<\/p>\n
However, as Meta pours hundreds of billions of dollars into data centers and high-priced chips, investors are growing increasingly anxious about when these massive capital outlays will yield returns. Commercializing computing power through cloud services not only provides Meta with a practical path to ease funding pressures and reassure investors but also marks a potentially significant transformation of its business model.<\/p>\n
Currently, the development of this cloud infrastructure business is still in its early stages, and the company’s ultimate strategy could still be adjusted. However, Zuckerberg has previously signaled that the option of leveraging large-scale AI infrastructure investments to generate revenue is already “on the table.”<\/p>\n
The performance of industry benchmarks also confirms the enormous capacity of this market. After decades of platform building, AWS, Azure, and Google Cloud have grown into massive businesses generating tens of billions of dollars in quarterly revenue by renting out computing power, storage, and software over the internet, providing a clear reference for Meta’s plans to commercialize its excess computing power. If Meta successfully enters this market, it could not only open up new profit models for itself but also potentially reshape the cloud computing landscape currently dominated by a handful of giants.<\/p>\n","protected":false},"excerpt":{"rendered":"Social media giant Meta Platforms (META) is brewing a major business transformation. Following reports that Alphabet (GOOGL) subsidiary…\n","protected":false},"author":2,"featured_media":92172,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[9],"tags":[24,1483,322,945,2408,132,1429,2194,48143,1123,4528,1124,48452,28854],"class_list":["post-92171","post","type-post","status-publish","format-standard","has-post-thumbnail","category-google","tag-ai","tag-alphabet","tag-aws","tag-coreweave","tag-gemini","tag-google","tag-google-ai","tag-google-cloud","tag-matt-bryson","tag-meta-platforms","tag-microsoft-azure","tag-muse-spark","tag-santosh-janardhan","tag-wedbush-securities"],"_links":{"self":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/92171","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/comments?post=92171"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/92171\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media\/92172"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media?parent=92171"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/categories?post=92171"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/tags?post=92171"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}