Meta Platforms Inc. (META) has quietly emerged as one of Microsoft Corp.’s (MSFT) largest artificial intelligence customers, spending hundreds of millions of dollars annually to access AI models through the Azure cloud platform, according to a Bloomberg report on Thursday. The arrangement underscores a persistent reality in the AI boom: demand for the technology remains overwhelmingly concentrated within the tech industry itself.<\/p>\n
People familiar with the matter said Meta consumes trillions of tokens — the basic unit of AI computational workload — through Azure every week. The company uses the service primarily to support its software development efforts, with developers tapping external models to build applications and benchmark the performance of Meta’s own in-house AI systems.<\/p>\n
The relationship is notable partly because Meta is one of the world’s most aggressive builders of AI infrastructure. The company expects to spend between $130 billion and $145 billion on capital expenditures in 2026, largely on AI and data center capacity. Yet it still rents third-party AI tools when doing so makes sense from an availability or cost standpoint, a hybrid strategy that Meta Chief Technology Officer Andrew Bosworth discussed publicly in a July podcast appearance.<\/p>\n
Microsoft’s AI Customer Base Remains Tech-Heavy<\/p>\n
Microsoft’s AI strategy centers on Azure Foundry, a model marketplace that offers AI models from multiple vendors rather than relying on a single partner. By July, Foundry had amassed roughly 100,000 customers, and Microsoft’s promotional materials frequently highlight deployments in traditional industries such as manufacturing and transportation.<\/p>\n
But behind that broad customer base, the largest spenders remain overwhelmingly tech companies. In addition to Meta, major Foundry customers include Adobe, AI search company Perplexity, and Sierra, a customer-service AI startup co-founded by OpenAI Chairman Bret Taylor. ByteDance, the Chinese parent of TikTok, has historically ranked as the largest paying customer on the platform overall, according to earlier reporting.<\/p>\n
Microsoft’s AI revenue is also heavily concentrated. OpenAI, the company’s long-standing strategic partner, accounted for approximately 70% of Microsoft’s overall AI revenue in its most recent fiscal year. Beyond model access, Microsoft also sells its Copilot AI assistant and rents out AI-optimized computing resources.<\/p>\n
CustomerRelationship to Microsoft AIOpenAIStrategic partner; ~70% of Microsoft AI revenueByteDanceLargest paying customer on Azure FoundryMeta PlatformsHundreds of millions annually; trillions of tokens weeklyAdobeMajor Foundry customerPerplexityMajor Foundry customerSierraMajor Foundry customer; co-founded by OpenAI Chairman Bret Taylor<\/p>\n
Note: Customer rankings reflect spending levels reported by people familiar with the matter and may not correspond to Microsoft’s official disclosures.<\/p>\n
The Concentration Problem<\/p>\n
Revenue concentration and so-called circular transactions — where tech companies buy AI services from one another — have long troubled industry observers. For AI to meet market expectations, the technology must achieve broad adoption across the wider economy rather than depending primarily on a handful of cash-rich technology firms.<\/p>\n
Analysts note that while Meta’s spending is significant for a single customer, it remains modest in the context of Microsoft’s overall cloud business. Azure revenue surpassed $100 billion in Microsoft’s fiscal 2026. A hypothetical $500 million annual contribution from Meta — the rough midpoint of the reported range — would equate to about 0.5% of Azure revenue. For Meta, the same figure represents roughly 0.4% of the midpoint of its 2026 capital expenditure guidance.<\/p>\n
A Pattern of Dependence and Replacement<\/p>\n
Meta’s reliance on Microsoft has historical precedent. In the late 2000s, Facebook used Microsoft’s Bing search engine to power web search functionality on the social network. By the end of 2014, Meta had fully discontinued its use of Bing, replacing the service with its own capabilities.<\/p>\n
Similarly, Microsoft provided Meta with computational infrastructure for AI research in the years before ChatGPT’s launch ignited the current AI boom. Since then, Meta has transformed into one of the largest builders of AI data centers globally.<\/p>\n
Meta is now developing its own API service to sell access to multiple AI models, a business that would compete directly with Azure Foundry. If successful, the initiative could help Meta reduce its dependence on external providers such as Microsoft.<\/p>\n
Meta developers have also used OpenAI models through Foundry to evaluate the output quality of Meta’s proprietary models, according to one person familiar with the arrangement. The company purchases model access across multiple platforms, weighing factors such as availability and cost.<\/p>\n
For Microsoft, Meta’s growing expenditure provides another marquee client and validates the company’s multi-vendor approach through Foundry. It also demonstrates that even the largest AI developers are willing to buy computing power and model access from rivals when it serves their needs.<\/p>\n
META shares closed at $546.03 on Wednesday, up slightly on the session. Neither company provided comment on the report.<\/p>\n","protected":false},"excerpt":{"rendered":"Meta Platforms Inc. (META) has quietly emerged as one of Microsoft Corp.’s (MSFT) largest artificial intelligence customers, spending…\n","protected":false},"author":2,"featured_media":146479,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[11],"tags":[755,8955,420,7829,34698,11373,6657,1123,320,7828,157,724,11375],"class_list":["post-146478","post","type-post","status-publish","format-standard","has-post-thumbnail","category-microsoft","tag-adobe","tag-andrew-bosworth","tag-azure","tag-azure-ai","tag-azure-foundry","tag-bret-taylor","tag-bytedance","tag-meta-platforms","tag-microsoft","tag-microsoft-ai","tag-openai","tag-perplexity","tag-sierra"],"_links":{"self":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/146478","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=146478"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/146478\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media\/146479"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media?parent=146478"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/categories?post=146478"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/tags?post=146478"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}