Google, Meta, Microsoft, Nvidia and Amazon all have the same 'AI problem', but why it is only troubling Mark Zuckerberg

In the midst of an unprecedented AI buildout, Meta is spending more than most. The company is building two massive data centers, and reporting indicates there will be as much as $600 billion in spending on U.S. infrastructure over the next three years. While such figures might be expected in the heart of Silicon Valley, they are beginning to raise alarms on Wall Street.

The concern hit a peak this week as Meta reported its third quarter earnings. The results revealed a substantial year-over-year jump in operating expenses, coupled with a staggering nearly $20 billion in Capital Expenditure (CapEx). This spending spree is driven by intense investment in AI talent and the necessary infrastructure, none of which has yet translated into meaningful, directly attributable revenue. When pressed for details by analysts, CEO Mark Zuckerberg confirmed that this spending is only the beginning.

Google, Meta, Microsoft, Nvidia and Amazon all have the 'same problem'

Meta isn't the only tech titan sinking billions into AI infrastructure. However, the market reaction to their spending is noticeably different compared to that of rivals like Google and Nvidia, both of whom recently posted strong quarters. Even OpenAI, arguably the biggest catalyst for this investment wave, spends comparable amounts despite having a far smaller financial cushion than Meta.

The core issue fueling investor anxiety is the fear of an AI bubble. If one forms, the consensus suggests Meta’s strong core business might allow it to weather the storm better than most, but the underlying question remains: where is the immediate return?

While specific recent quarterly CapEx figures fluctuate, reports often show that Google, Microsoft, and Amazon are all substantially increasing their CapEx, albeit often in ways more directly tied to immediate, established revenue streams like Cloud services (Microsoft Azure, Google Cloud, Amazon Web Services) or dominant advertising/search platforms. Sam Altman of OpenAI justifies his hundreds of billions in compute spending by pointing to one of the fastest-growing consumer services in history, backed by a reported $20 billion annual revenue run rate (ARR). That rapid growth offers a clear, quantifiable answer to sustainability questions.

Meta's missing product

Meta’s spend, by contrast, is being directed toward infrastructure whose direct monetization path is less clear in the immediate future. Meta currently lacks a comparable, runaway AI product. Its most visible AI offering is the Meta AI assistant, which Zuckerberg noted has over a billion active users. However, this figure is heavily "juiced" by the three billion active users across Facebook and Instagram, and the current version struggles to stand as a true competitor to specialized models like ChatGPT. Furthermore, features like the Vibes video generator successfully boost daily active users but have limited long-term business impact. Vibes is a platform where users can create and share short-form, AI-generated videos.

The most ambitious physical product, the Vanguard smart glasses, feels more like an extension of Meta’s established Reality Labs efforts than a direct, large-scale application harnessing the raw power of Large Language Models (LLMs) for immediate, scalable profit.

In short, Meta is making monumental bets on infrastructure that is currently funding promising experiments, not yet fully formed, direct revenue-generating products.