quick read
-
meta platform (META) plans to spend $145 billion in capital spending on AI infrastructure in 2026, outpacing the spending rates of Microsoft, Alphabet, and Amazon, while CEO Mark Zuckerberg has hinted that the company may launch a cloud computing business to monetize excess capacity and compete with Amazon Web Services, Google Cloud, and Microsoft Azure. Meta generated $45 billion in free cash flow over the past four quarters, giving it the financial ability to fund expansion without weakening its balance sheet.
-
After spending $70 billion on Metaverse with limited revenue, Meta is transforming from a social media company to an AI hyperscaler, betting that AI infrastructure will become a profitable platform business, similar to the emergence of cloud computing a decade ago.
-
Act now. Analysts who called NVIDIA in 2010 just named it their top 10 AI stocks. And Meta was not selected. Get your name for free now.
The artificial intelligence arms race has turned Silicon Valley into a spending race. meta platform (NASDAQ:Meta), microsoft (NASDAQ:MSFT), alphabet (NASDAQ:GOOG), and Amazon (NASDAQ:AMZN) is pouring hundreds of billions of dollars into chips, servers, networking equipment, and data centers in a race to dominate AI infrastructure. The problem investors are grappling with is simple. How much is too much to spend?
For Meta shareholders, the concerns are only growing stronger. CEO Mark Zuckerberg told investors at Meta’s annual shareholder meeting on Wednesday that he expects the company’s capital spending to reach $145 billion in 2026. This number dwarfs the company’s already huge spending plans and solidifies Meta’s transformation from a social media company to a full-fledged AI hyperscaler.
But surprisingly, Zuckerberg may also have revealed a safety valve in case Meta builds up too much capacity.
Act now: Analysts who called NVIDIA in 2010 just named their top 10 AI stocks, and Meta wasn’t on the list. Get your name for free now.
Meta’s AI ambitions extend beyond the Metaverse
Just three years ago, Meta became synonymous with the Metaverse. The company spent tens of billions of dollars through its Reality Labs division pursuing VR adoption, but it never fully materialized, generating cumulative operating losses of more than $70 billion between 2021 and 2025.
Now the focus has shifted almost entirely to AI. However, unlike the Metaverse push, Meta’s AI spending is tied directly to its core business. AI recommendation engines are already powering content discovery across Facebook, Instagram, and Threads. Advertising tools powered by generative AI have improved campaign targeting and increased engagement metrics feeding into Meta’s ad machine. It still accounts for about 97% of revenue.
Let’s put the scale of this expenditure into perspective.
|
company
|
Estimated capital investment in 2026
|
main focus
|
|
meta platform
|
$125 billion to $145 billion
|
Internal AI infrastructure
|
|
microsoft
|
Approximately $190 billion
|
Azure + OpenAI infrastructure
|
|
alphabet
|
$175 billion to $190 billion
|
Google Cloud + Gemini AI
|
|
Amazon
|
Approximately $200 billion
|
Extending AWS
|
The main difference is that Meta doesn’t primarily build cloud infrastructure for external customers. At least not yet.
This distinction is important because investors are concerned that the meta will repeat the same pattern seen during the Metaverse: massive spending with no clear return on investment.
Zuckerberg’s backup plan changes the equation
This concern helps explain why Mr. Zuckerberg’s comments at the shareholder meeting turned out differently than past spending announcements.
Zuckerberg said that if Meta builds out excess AI capacity, launching a cloud computing business is “definitely on the table.” He noted that third parties regularly approach Meta for access to Meta’s computing power and are willing to pay a premium price to do so.
So Meta may be building an escape hatch into one of the tech industry’s most profitable businesses. That would put Meta in direct competition with the following companies:
-
Amazon Web Services
-
google cloud
-
microsoft azure
Granted, these businesses took years to scale. Amazon Web Services had $128.7 billion in revenue last year, while Microsoft’s intelligent cloud division had annual revenue of more than $131 billion. Still, Meta has a history of being late to established markets and succeeding anyway.
Instagram Stories has copied Snapchat. Reels copied TikTok. The thread appeared after the confusion over X. Meta rarely invents this category, but often wins through scale, engineering talent, and distribution. Cloud infrastructure businesses will follow the same strategy.
There’s good reason for investors to be nervous, but there’s also opportunity.
That said, the skepticism is understandable. Meta stock fell more than 75% from late 2021 to late 2022 as investors questioned Zuckerberg’s Metaverse spending. The market hasn’t forgotten that lesson.
At current levels, Meta is trading at a premium valuation compared to its historical average, as investors expect AI investments to generate real earnings growth. If spending accelerates faster than revenue, profit margins may shrink and stock prices may come under pressure.
In any case, the meta position is stronger today than it was during the construction of the metaverse. The company generated more than $45 billion in free cash flow over the past four quarters, giving it significant financial strength to fund its AI expansion without hurting its balance sheet.
Zuckerberg may be making a calculated bet that AI infrastructure itself will become the next platform business, much like cloud computing was a decade ago.
Important points
Meta’s $145 billion AI gamble comes with real risks. Sharp investors shouldn’t ignore the possibility that its expenses could outweigh its revenues, especially after the company’s costly detour into the Metaverse.
But Zuckerberg’s comments revealed something investors may have overlooked: Meta isn’t just building AI infrastructure for experimentation. You may be building the foundation for an entirely new line of business.
If Meta eventually monetizes its excess capacity through cloud services, today’s spending may look less like reckless expansion and more like the early stages of another technology empire. For long-term shareholders, that possibility completely changes the conversation.
Act now: Analysts who called NVIDIA in 2010 just named their top 10 AI stocks, and Meta wasn’t on the list. Get your name for free now.
