Meta $META Margins Crater to 31%: Is the AI Spending Spree Worth It?
Meta beat on revenue but Wall Street is asking a harder question: when does all this AI capex actually pay off?

Ticker Ratings
$META just proved you can grow revenue 28% year-over-year and still get punished by the market. After-hours trading slapped the stock down roughly 6-9% depending on which Bloomberg anchor you were watching, because the number that matters right now is not the top line. It is the operating margin, which collapsed from 43% to 31%. That is not a rounding error. That is a company setting money on fire at industrial scale.
The culprits are not mysterious. A $2.4 billion legal charge, $1.2 billion in severance costs, and surging depreciation from a capex cycle that keeps getting bigger. Meta raised the lower bound of its full-year capex range to $230 billion, and also bumped up the floor on total expenses. The Q3 revenue guidance midpoint came in below analyst estimates, which is the kind of thing that turns a mildly disappointing print into a genuine selloff.
The bull case is not dead, but it needs to be honest about what it is. Azure just printed 43% cloud growth and $100 billion in annual revenue for $MSFT, proving that AI infrastructure spending can translate into real revenue acceleration. Meta's version of that story, Copilot-style monetization through Advantage Plus ad tools and AI-generated creative, is still more theoretical than proven. Zuckerberg is betting that the margin pain is temporary and the revenue flywheel kicks in by 2027. The market is betting he is wrong, at least for now.
The broader context makes this more interesting. We are in a week where the Philadelphia Semiconductor Index dropped 5.3% in a single session and nearly 16% over five trading days, hyperscaler capex sustainability is the dominant earnings narrative, and even $QCOM missed on EPS and guided light for Q3. The question the street is asking across every big tech print is the same one: are these companies spending into a moat or spending into a margin trap?
For Meta specifically, the social sentiment on YouTube and X is split down the middle. The bears are screaming about cost discipline. The bulls are pointing to 30 million Copilot seats at Microsoft as a proof point that AI monetization is real and Meta will find its version of it. Both are right in different time frames, which is the worst possible outcome for anyone trying to trade the next two weeks.
Somewhere in Menlo Park, Zuckerberg is building the metaverse of AI models and hoping the ad business keeps funding the party long enough for it to matter.