Meta's stock remains one of 2026's worst performers, down more than 17% year-to-date. Despite a flurry of AI announcements this month—including new smart glasses, enterprise tools, prediction-market apps, and a Qualcomm partnership to boost computing power—Wall Street continues to penalize the company. The core issue: investors are no longer impressed by AI features; they want proof that the $125–145 billion capex spend actually generates revenue.
Meta raised its 2026 capex guidance by $10 billion (midpoint) in Q1 earnings due to rising costs for memory, chips, and data-center components. That guidance hike tanked shares 9% on the report. Analysts now frame the math simply: Meta's free cash flow is "basically going to zero" as capex consumes nearly all net income. The Magnificent Seven splits are stark: only Microsoft (down 26%) has underperformed Meta among the giants, and neither possesses a high-margin cloud business (unlike Alphabet, Amazon, Microsoft) to justify heavy AI spend.
For practitioners funding or evaluating AI infrastructure: Meta's struggle underlines a hard reality. Building your own models, data centers, and silicon is a $100B+ per-year commitment with a 2–3 year horizon to ROI visibility. Cloud incumbents and hyperscalers face increasing pressure to show near-term monetization, not just long-term optionality. Teams considering in-house AI R&D budgets should model cash-flow impact and shareholder tolerance for prolonged capex before committing.