Meta’s AI story just got more credible, and that makes the bull case harder to dismiss. The key shift is simple: this is no longer just a company talking about AI ambition, but one planning to put an in-house AI chip into production in September while targeting 14 gigawatts of compute next year. That matters because Meta already has the scale to turn infrastructure advantages into product gains across ads, messaging, and consumer AI. At 23.93 times trailing earnings, this still does not look priced like a business with fake momentum.
The first reason to stay bullish is that Meta is funding this AI build from a position of real operating strength, not hope. Revenue grew 22.2% year over year to $200.97 billion, while operating margin held at a massive 41.2% and net margin came in at 32.8%. Those are not the numbers of a company losing control of spending. They are the numbers of a company with enough earnings power to absorb heavy infrastructure investment and still print elite profitability, which is exactly why the TickerSpark Score gives Meta a perfect 100 on Profitability and 96 on Financial Health.
The second reason is that the market is finally getting a tangible execution marker instead of a vague AI narrative. September production for Meta’s in-house AI chip, a roadmap of four in-house chips, and a plan to launch a new chip roughly every six months through 2027 all point to a real silicon program, not a science project. If Meta reaches its stated 14-gigawatt compute target next year, the debate shifts from whether capex is reckless to whether Meta is building one of the deepest proprietary AI stacks in large-cap tech. That is exactly the kind of infrastructure advantage that can widen moats over time.
The third reason is that the stock still has room to work even after the latest jump. META closed at $669.21 after a 6.0% move, but the valuation is still below Alphabet on a growth-adjusted narrative that is getting stronger, with Meta at 23.93 times earnings versus GOOGL at 27.24. Meanwhile, Meta’s revenue growth of 22.2% is running well ahead of Alphabet’s 15.1%. The TickerSpark Score overall sits at 78, with Momentum at 80, and the tape agrees: shares are above the 50-day and 200-day moving averages, MACD is positive, and on-balance volume is in accumulation. This does not look like a broken chart being rescued by hype.
The fourth reason is that execution credibility keeps showing up in the numbers that matter to the Street. Meta has beaten earnings estimates in 7 straight reported quarters, including a 7.2% beat in April and 8.0% in January. Consensus still leans bullish, with 50 Buy ratings against 11 Holds and just 2 Sells, and recent news sentiment is strongly positive with a 7-day reading of 0.8347. That does not guarantee upside, but it does show that the market is rewarding delivery, not just storytelling.
The cleanest pushback is that none of this proves near-term monetization. The chip program may first lower GPU costs and reduce outside dependence rather than unlock a visible revenue stream, and Meta’s year-over-year EPS growth is currently down 2.6% while net income growth is down 3.1%. That is a fair warning that the spending cycle is still pressuring the bottom line even with revenue growing fast.
There is also some real execution risk in the silicon roadmap, especially because training-chip ambitions have been tougher than inference chips. Insider activity does not help the optics either, with 10 recent sells totaling $3.30 million and no buys. Even so, that counterargument falls short because Meta is not being asked to prove a brand-new business from scratch. It is being asked to make its existing ad and AI ecosystem more efficient and more defensible, and few companies are starting from a base of 81.9% gross margin and $60.46 billion in net income.
That leaves META looking like a stock to own on strength, not fade on headlines. We would respect the recent breakout rather than fight it, especially with shares still above the 200-day moving average of $642.72 and sentiment improving into the next earnings report. The setup only really breaks if the chip timeline slips or if management fails to show that this capex wave is supporting margins, engagement, or AI product adoption.
For now, the burden of proof has shifted in Meta’s favor. The AI chip move does not finish the monetization story, but it makes the infrastructure story real enough that dismissing the bull case now means arguing against both execution momentum and one of the strongest profit engines in the market.
