HUGE: Meta’s Ad Empire Is Fueling a $145B AI Bet

Added: 2026-05-07

[00:00:00]Google is scaling AI through search.

[00:00:03]Correct? Amazon is turning cloud into an AI machine. Microsoft is building the agent stack. You were wondering about Meta, right? So was I. And then, of course, there's Meta. Beating expectations across the board and still getting punished by the market because while everyone is talking about AI models, Zuckerberg is doing something far more aggressive. He's rebuilding Meta as a super intelligence company. Meta just reported its Q1 2026 earnings. Revenue came in at $56.3 billion, which is up 33% year-on-year. EPS beat expectations.

[00:00:46]The Q2 guidance looks quite solid and yet the stock dropped after hours. Why?

[00:00:53]Well, because Meta raised its CapEx guidance again from $115 $115 to $135 billion to $125 to $145 billion. Mark Zuckerberg called this a milestone quarter. He said, "Meta has already released its first model from Meta super intelligence labs and is now on track to deliver personal super intelligence to billions of people." End quote. Investors are not questioning Meta's business so much.

[00:01:25]They're actually questioning the spending because this is what actually the numbers look like. CapEx is $19.8 billion this quarter. The full year projection is up to $145 billion.

[00:01:36]Costs are growing faster than revenue, which is a matter of concern. User growth is slowing slightly quarter-over-quarter due to, well, the geopolitical disruptions. And for the market, this raises one significant concern. Is Meta overspending in the AI race? Meta generated $12.4 billion in free cash flow in a very in in a single quarter and is right now sitting on over $81 billion in cash. And the reason is, well, Meta is capacity constrained.

[00:02:08]Even after years of building data centers, it still doesn't have enough compute. Every signal inside Meta is pointing in one direction, which is compute demand is exploding. AI models are getting bigger. Ad systems becoming LLM driven. Personal and business agents scaling globally. Infrastructure commitments jumping by a hundred plus billion dollars in a single quarter.

[00:02:35]Even internally, Meta admits it has consistently underestimated how much compute it needs. And this is where Meta is different. Google is using AI to improve search. Microsoft is embedding AI into enterprise workflows. Amazon is monetizing AI through cloud infrastructure. Meta Meta is doing something else entirely. It is building consumer AI at global scale, personal agents for billions, business agents across WhatsApp, Instagram, Messenger, AI-powered ads engine driving revenue, all on top of its own infrastructure.

[00:03:09]And here's the real edge. Meta doesn't need external customers to justify its AI spend. Why? Well, because its ads business is already funding everything.

[00:03:20]$55 billion plus ad revenue in a single quarter. Wow. AI improving conversion rates and monetization. 8 million plus advertisers using gen AI tools.

[00:03:32]AI-driven engagement pushing ad performance higher.

[00:03:37]But, of course, there are cracks. Super intelligence claims have, well, no external validation as yet. User growth is now sensitive to geopolitics. Reality Labs continues to burn billions.

[00:03:52]Regulatory pressure is building globally. And the biggest one is Meta is trading on future AI credibility, not current fundamentals. So, the real question is, is Meta overspending or is it just moving faster than the market can truly understand? Please do let us know what are your thoughts in the comments below. This, ladies and gentlemen, is Front Page by the I AM Network. Like, share, subscribe, and always remember, yes, you said it. Think AI. Think I AM.