The market has already decided what it thinks about Big Tech AI spending. Alphabet raised its capex forecast last week and saw shares fall more than 7% the same day. Tonight, Meta gets its turn.
Meta Platforms reports Q2 2026 earnings after the close tonight, and the setup is genuinely interesting. Not because a revenue beat is in doubt. It almost certainly isn’t. The real question is whether 3.5 billion daily users and a 27% revenue growth rate are enough to absorb what is shaping up to be the most aggressive capital spending program in the company’s history.
Revenue consensus sits at roughly $60.2 billion, up about 27% year over year. Meta guided the quarter to a range of $58 billion to $61 billion, and the Street has anchored near the top of that range. The underlying ad business is strong. Q1 saw a 19% jump in ad impressions with average price per ad climbing 12%. That combination does not crack easily.
But here is where it gets complicated.
Meta has beaten estimates in four consecutive quarters, posting an average EPS surprise of about 12%. And yet the stock has traded down in 2026 anyway. It is off roughly 9% year to date and sitting about 25% below its 52-week high of $796.25. The market is not punishing a failing business. It is repricing a business that may be spending faster than it can prove returns.
Full-year 2026 capex guidance already sits at $125 billion to $145 billion, raised from $115 billion to $135 billion last quarter. Wedbush projects Meta will spend approximately $138 billion this year, with one model projecting 2027 capex rising toward $173 billion. For context, operating margin is expected to fall to roughly 33.8% from 43% a year earlier, even as revenue grows.
The part people skip: Meta is currently the only major hyperscaler without a cloud business it can point to as proof its AI spending earns money outside of advertising. Alphabet has Google Cloud. Amazon has AWS. Microsoft has Azure. Meta has an ad platform and a Reality Labs division still losing roughly $5 billion per quarter. Any commentary on the reported Anthropic cloud talks or a monetization path for its AI infrastructure would change the calculus meaningfully.
The options market is pricing in about an 8% implied post-earnings move. That is elevated. And Meta has beaten the implied move in six of its last eight reports.
The bull case is real. A stock at roughly 18 to 22 times forward earnings, below its own three-year historical average of about 23 times, with close to 27% top-line growth and ad pricing power that has not broken, is not an obvious short. If capex guidance holds at the existing range and management signals that the spending pace stabilizes in 2027, the stock could close a meaningful portion of the gap to its high.
The risk is the opposite: another guidance raise on infrastructure spending, another quarter where EPS growth barely moves while revenue accelerates, and a market that has already shown it will sell the beat if the capex number disappoints.
What happens tonight matters beyond Meta. Microsoft reports tonight as well, and between the two, the market gets its clearest read yet on whether the AI capex supercycle is being rewarded or penalized. That answer will set the tone for technology into August.
The full breakdown drops after the close. Keep an eye on the call at 4:30 p.m. ET. That is where the guidance that actually moves the stock gets delivered.
