Meta Posts Record Q1 Revenue Up 33%, but $145B Capex Ceiling Sends Stock Down 5%

Meta’s Q1 2026 earnings report told two stories simultaneously — and the market, at least in the immediate aftermath, chose to focus on the more uncomfortable one.

The company reported revenue of $56.31 billion, up 33% year-over-year, beating the $55 billion analyst consensus and delivering earnings per share of $10.44. On almost any surface reading, those are exceptional numbers for a company Meta’s size. The AI-driven advertising thesis — that Meta’s Llama models are generating improvements in ad targeting significant enough to drive revenue per impression higher — continues to deliver measurable results.

But Meta also raised its 2026 capital expenditure guidance from a range of $120-$135 billion to $125-$145 billion, citing higher component prices and additional data center investment. That $10 billion increase at the top end of the range was enough to send shares down more than 5% in after-hours trading, even as the revenue beat confirmed that AI monetization is working.

Revenue Beat: AI Ads Are Delivering

The Q1 revenue of $56.31 billion represents 33% year-over-year growth — an impressive rate for a company generating more than $200 billion in annual revenue at run rate. The performance was driven by a combination of factors: continued growth in the Family of Apps daily active users, improvements in ad load and monetization efficiency, and most significantly, the ongoing impact of AI-powered ad ranking and targeting systems built on Meta’s Llama foundation models.

Meta has been one of the clearest proponents of the thesis that large language models can materially improve advertising performance through better understanding of user intent, creative optimization, and precision targeting. The Q1 results are another data point confirming that thesis: revenue per impression continues to grow, and Meta’s ad products are gaining share against digital advertising alternatives.

Net income came in at $26.8 billion, or $10.44 per diluted share. However, the reported EPS includes a significant one-time benefit: an $8.03 billion income tax benefit related to the One Big Beautiful Bill Act and a U.S. Treasury notice that altered the tax treatment of certain international operations. Stripping out this tax benefit, net income was approximately $18.7 billion and adjusted EPS would have been $7.31 — still a strong result, but the headline number requires this important asterisk.

Q2 Guidance: Flat, With Currency Headwinds

The Q2 2026 revenue guidance range of $58 billion to $61 billion implies growth in the mid-to-high twenties percent range, roughly in line with Q1 — assuming the midpoint. Management noted that foreign currency is an approximately 2% tailwind to year-over-year revenue growth in the quarter, suggesting that the underlying constant-currency growth rate is slightly lower than the headline figure.

The guidance was read as conservative by some analysts, particularly given Q1’s strong performance and the sequential revenue step-up implied by the midpoint versus Q1 actuals. Whether this represents true conservatism or a signal of macro uncertainty — particularly around advertising spending trends in markets affected by tariff policy — will be a topic for the next quarter’s call.

The Capex Shock: $145 Billion and Counting

The market-moving element of Meta’s report was the capex revision. The company now expects 2026 capital expenditure of $125 billion to $145 billion, up from the prior range of $120 billion to $135 billion. CFO Susan Li cited two factors: higher pricing on data center components (a reflection of AI chip supply constraints across the industry) and accelerated investment in AI infrastructure capacity.

The $145 billion ceiling is striking in its own right. It exceeds the annual GDP of many mid-sized countries and represents approximately 25% of Meta’s projected 2026 revenue being reinvested in infrastructure. For a company that generated its historical returns largely through relatively capital-light social media infrastructure, the transformation in capital intensity is jarring.

The investor reaction — stock down 5% despite a revenue beat — reflects a specific anxiety: Meta’s spending is increasing faster than its revenue guidance implies returns will arrive. If the incremental capex generates AI capabilities that translate into future revenue, the spending is justified. If it doesn’t — or if it does, but on a three-to-five-year rather than one-to-two-year timeline — the near-term EPS compression becomes a problem.

Zuckerberg addressed this directly on the earnings call. “We are in a multi-year transition to a future where AI is central to every product and every user interaction across our platforms,” he said. “The infrastructure we’re building is not optional infrastructure — it is the substrate on which our next decade of growth will run.” He specifically described the capex increase as a response to “exceptional demand” signals across Meta’s AI products and services.

Superintelligence Labs: The Long-Term Bet

The earnings call also featured discussion of Meta’s Superintelligence Labs, the organization Zuckerberg formally established earlier in 2026 with the stated mission of building artificial general intelligence. Meta recruited dozens of top AI researchers from DeepMind, OpenAI, and academic institutions for the initiative, and the Labs have been operating with unusual independence within Meta’s organizational structure.

Analysts probed whether the Superintelligence Labs spending is included in the $145 billion capex ceiling and how it interacts with Meta’s near-term product roadmap. Zuckerberg was characteristically non-specific about timelines but clear about intent: “We believe the race to AGI is the defining competition of our generation, and Meta is fully committed to competing for the frontier.”

The tension embedded in that statement is real. Meta’s near-term financial performance depends on AI improving its advertising products — a high-confidence, high-return deployment of AI capability. The Superintelligence Labs moonshot is a different kind of bet: lower certainty, longer timeline, potentially enormous upside, but not generating revenue next quarter or the one after. Investors pricing Meta on near-term earnings multiples are essentially being asked to trust that the long-range bet doesn’t crowd out the near-term monetization engine.

Weighing the Quarter

Meta’s Q1 2026 results are, on balance, strong. A 33% revenue growth rate, continued AI advertising momentum, and a product roadmap that is visibly executing confirm that the company’s core business remains healthy and that AI is generating measurable improvements in its primary revenue stream.

The capex revision is the legitimate concern. At $145 billion, Meta is making a larger infrastructure bet than almost any company in history, and it is making that bet against a backdrop of foreign exchange headwinds, macroeconomic uncertainty, and management’s own admission that the returns from some of this spending are measured in years rather than quarters.

The market’s 5% post-earnings decline is less a verdict on the quarter’s results than a pricing-in of that uncertainty. For long-term holders of Meta, the relevant question is whether Zuckerberg’s conviction that AI infrastructure is a “generational” investment will prove correct. The Q1 numbers suggest the near-term fundamentals support the bet. Whether the $145 billion ceiling is the right number — or simply the beginning of an even larger one — is the question that will define Meta’s investment narrative for the rest of the decade.