Meta Just Crushed Earnings. So Why Does Wall Street Keep Selling the Stock?

The market is treating Meta’s AI spend as a pure cost item, but the more important read-through is that the capex is now acting like a moat-expansion tax on everyone else. If Meta sustains ad pricing and impression growth while funding inference/training infrastructure at this scale, smaller ad-tech and social competitors will face a widening quality gap in targeting and monetization that they cannot match with incremental budgets. That makes the near-term winner not the consumer AI products, but the core ad engine — a business that can self-fund aggressive reinvestment faster than rivals can respond.

The real risk is timing mismatch. Investors are discounting a multi-year payback window using a near-term margin lens, and that can stay painful for several quarters if capex guides up again or if Reality Labs losses re-accelerate. The stock’s sensitivity is now less about headline earnings and more about whether management can show that each incremental dollar of spend improves ad load, CPC, or engagement faster than depreciation and operating expense growth.

Contrarian view: the consensus is underestimating how much optionality Meta is buying with a still-cheap multiple relative to its earnings power. If AI tools lift ad ROI even modestly, the market will eventually re-rate Meta as a cash-generating distribution layer with embedded AI software economics, not a speculative hardware/consumer-AI story. The inflection point is likely in the next 2-3 quarters: if ad metrics remain elevated while capex is absorbed without margin collapse, the current drawdown could prove to be an entry point rather than the start of a de-rate.

Second-order winner/loser dynamics also matter: AI infrastructure vendors, networking, and power/thermal suppliers benefit from Meta’s spending cadence, while any ad-tech names relying on weaker targeting or lower-intent inventory get squeezed. The stock is vulnerable only if the market starts to believe capex is chasing prestige projects rather than measurable ad yield — that would flip sentiment fast and create another leg down.