Remember when everyone, and I mean everyone, was freaking out about Meta’s spending? The narrative was almost a meme: “Zuck is burning billions on the metaverse and AI with nothing to show for it!” It felt like every financial news headline was a variation of that same theme. Investors were nervous, analysts were skeptical, and the general vibe was that the company was on a wildly expensive ego trip.
Well, after Meta dropped its Q2 2025 earnings report, that entire story got shredded and tossed into the digital dumpster. It’s one of the most stunning narrative flips I’ve seen in tech. The skeptics are suddenly believers, and the conversation has shifted from “reckless spending” to “genius-level investment.”
Meta’s stock didn’t just nudge up; it rocketed over 11% in a single day. That’s not just a good day; that’s a massive vote of confidence from the market, a collective realization that the strategy is working. And it wasn’t just retail investors getting excited. The big Wall Street firms, including Rosenblatt, Bank of America, and Morgan Stanley, are all scrambling to raise their price targets, basically admitting they underestimated the beast Meta was building.
✨ The AI Engine Roars to Life
So what caused this dramatic 180? It’s all about the AI. For years, Meta has been pouring an eye-watering amount of cash into what they call Capex, or capital expenditures. Think of Capex as the money spent on the heavy-duty hardware: the servers, the custom AI chips, the data centers, that forms the backbone of their empire.
Previously, this spending was seen as a black hole. Now, it’s viewed as the fuel for an incredibly powerful growth engine. The returns are no longer theoretical; they’re showing up in the numbers, loud and clear. It’s a classic AI flywheel effect, and it’s spinning faster than anyone expected.
Here’s how that flywheel is crushing it:
- 📌 Resurgent Ad Growth: This is the big one. Ad growth is back to over 20% on a jaw-dropping $160 billion base. That’s like turning an aircraft carrier on a dime. The AI is making ads so much more effective that advertisers are getting better results and throwing more money at the platform.
- ✅ Supercharged Conversions: Meta’s AI isn’t just showing ads; it’s predicting exactly who is most likely to click and buy. This increases ad conversions, which is the holy grail for advertisers. Better conversions mean a higher return on ad spend (ROAS), making Meta’s platforms the best place to put your marketing dollars.
- 💡 Insane Engagement: The AI is also making Facebook and Instagram stickier. The recommendation algorithms are getting scarily good at showing you content that keeps you scrolling. More time on the app means more data for the AI to learn from and more opportunities to show you hyper-relevant ads.
- 🚀 Creative Tools for Everyone: It’s not just about consumption. Meta has rolled out AI video generation tools, and nearly 2 million advertisers are already using them. This is a game-changer, allowing small businesses to create high-quality video ads without a huge budget, further cementing their reliance on the Meta ecosystem.
⚙️ The Capex “Problem” Is Now a Superpower
Let’s talk about those spending numbers, because they are staggering. Rosenblatt analysts are now modeling Meta to spend $68 billion on capex in 2025 and a mind-blowing $97 billion in 2026. Bank of America expects a $30 billion increase in capex in 2026 alone.
A year ago, numbers like these would have sent investors running for the hills. Today, they’re cheering. Why? Because every billion dollars Meta spends on AI infrastructure is now directly correlated with revenue growth. The investment isn’t a cost anymore; it’s the price of admission to dominate the next era of technology.
This is a fundamental shift in understanding Meta’s business. They aren’t just a social media company anymore. They are an AI company that happens to own some of the world’s largest social platforms. The spending is what gives them their edge, the sheer scale of their computational power is something very few companies on Earth can even dream of matching.
And it’s not just about the core apps. Even their hardware bets are starting to pay off. The Ray-Ban smart glasses saw sales jump 200% in the first half of 2025. It’s a small piece of the puzzle, but it shows that Meta’s innovation engine is firing on multiple cylinders.
✍️ The Ultimate Goal: Personal Superintelligence
This is where it gets really exciting. All of this spending, all of this AI development, it’s all leading to one audacious, long-term goal: creating “Personal Superintelligence.”
This isn’t just a smarter Siri or a more helpful chatbot. Think bigger. Meta’s vision is an AI that acts as a true digital companion and assistant for your entire life. An AI that understands your context, anticipates your needs, helps you learn new skills, manages your communication, and empowers your creativity in ways we can barely fathom today.
This is the North Star that justifies the $100 billion spending sprees. They aren’t just building a better ad machine; they are in a race to build the next fundamental computing platform. The same way the PC and the smartphone changed everything, Meta believes a personal, superintelligent AI will be the next paradigm shift.
And now, Wall Street is finally starting to see it too. They realize that the company that wins this race will be unimaginably valuable. The narrative is no longer about the cost of the investment, but about the sheer size of the prize. Meta played the long game, endured the criticism, and now the strategy is being validated in the most powerful way possible: with incredible results.
- Massive Capital Investment: Meta’s commitment to AI is reflected in its spending. The company’s capital expenditures soared 101% year-over-year to $17.01 billion in the second quarter. It projects spending between $66 billion and $72 billion for the full year 2025 and expects a similar growth rate in 2026 to build out its AI infrastructure.
- Strategic Acquisitions and Hires: To lead its AI charge, Meta has not only invested $14.3 billion in Scale AI but has also recruited top industry talent. This includes former GitHub CEO Nat Friedman and ex-Scale AI CEO Alexandr Wang, who now help lead the company’s new “Superintelligence” unit.
- A Tale of Two Divisions: The financial results highlight a stark contrast within the company. While the Family of Apps (Facebook, Instagram, etc.) generated $46.6 billion in advertising revenue, the Reality Labs division, which houses the long-term AI and metaverse bets, reported a loss of $4.5 billion for the quarter.