Meta Platforms reported first-quarter 2026 earnings on Wednesday April 29 that beat Wall Street expectations on both revenue and profit. The stock fell 10 percent the following day anyway.
That apparent contradiction is the whole story of Meta’s relationship with Wall Street right now, and understanding it requires understanding what the company said about the money it plans to spend on artificial intelligence over the rest of the year.
Here is what happened, what the numbers say, and why the stock went in the wrong direction despite a genuine earnings beat.
Meta Beat Earnings And The Stock Fell Anyway
The Q1 numbers were strong by any conventional measure. Revenue came in at $56.31 billion against Wall Street estimates of $55.45 billion, a beat.
Adjusted earnings per share came in at $7.31 against the $6.79 analysts expected, a substantial beat.
The company’s advertising revenue, which is its core business, grew 33 percent year-over-year to $55.02 billion, the fastest quarterly growth since 2021. CEO Mark Zuckerberg said the company had “a milestone quarter with strong momentum across our apps.”
Two things went wrong. The first was user growth.
Daily Active People, the metric Meta uses to measure how many people use at least one of its apps on any given day, came in at 3.56 billion, up 4 percent year-over-year but below the 3.62 billion analysts had expected.
Meta attributed the sequential pullback to specific regional disruptions: internet blackouts in Iran connected to the ongoing war and blocked access to WhatsApp in Russia.
The second and more significant problem was what the company said about spending.
The $145 Billion Question
Meta raised its full-year 2026 capital expenditure guidance to a range of $125 billion to $145 billion.
The previous guidance had been $115 billion to $135 billion. The increase of $10 billion at both ends of the range was attributed to higher component pricing, the same explanation Microsoft gave the same night for its own capex increase, and additional data center spending to support future capacity.
Investors were already bracing for a large number. Analysts had been modeling approximately $122.6 billion in full-year capex.
The new guidance came in above even those elevated expectations at the top end.
The Q1 capex itself was $19.84 billion, actually below the $27.57 billion average estimate, but the full-year guidance is what matters to long-term investors evaluating when and whether the spending produces returns.
JPMorgan analysts downgraded the stock to Neutral from Overweight on Thursday, saying Meta faces a “challenging path” to generating returns on its heavy capex forecast.
They cited the company’s lack of enterprise tech stack integration, the deep institutional connections that make it harder for businesses to stop using a cloud provider once they have built on it, as a structural disadvantage compared to the hyperscalers who sell cloud services externally.
The core of their concern was simple. Meta’s AI spending is largely internal, building infrastructure for its own recommendation systems, generative advertising tools and Llama models.
That is investment in the cost side of the business, not the revenue side. There is no Google Cloud equivalent at Meta generating tens of billions in external revenue to justify the capex.
Zuckerberg has a vision for where this leads. “We’re on track to deliver personal superintelligence to billions of people,” he said in his statement.
Whether that vision produces the kind of revenue that justifies $145 billion in annual capital expenditure is the question Wall Street is asking, and the company has not yet answered it in ways the market finds satisfying.
The stock is being cut while the company is simultaneously being asked to do something unusual. Cut costs and raise spending at the same time.
Meta announced it is laying off approximately 10 percent of its workforce, 8,000 employees effective May 20, plus 6,000 open roles that will not be filled, while raising the capex forecast.
The layoffs are intended to free up resources for AI investment. The combination reads, to investors who are skeptical, as a company under pressure to show discipline while simultaneously making its biggest bet ever.
Why Google Rose In Contrast
The contrast between Meta’s Thursday and Alphabet’s Thursday tells the full story of what Wall Street is willing to pay for right now.
Alphabet reported the same night as Meta and also raised its capex guidance. Full-year 2026 capex went to $180 billion to $190 billion, up from $175 billion to $185 billion.
CFO Anat Ashkenazi said the company expects 2027 capex to “significantly increase” from 2026. That is a larger absolute number than Meta’s guidance, and the directional signal is the same — spending is going up.
The stock rose more than 5 percent on Thursday anyway. The reason is Google Cloud.
Google Cloud revenue in Q1 2026 came in at $20.03 billion against estimates of $18.05 billion, a beat of nearly $2 billion, or roughly 11 percent above expectations.
That represents 63 percent year-over-year growth, accelerating sharply from the 48 percent growth the division posted in Q4 2025.
The Cloud backlog, money customers have committed to spend with Google Cloud but have not yet spent, reached approximately $460 to $462 billion, having nearly doubled in a single quarter.
Sundar Pichai, Alphabet’s CEO, made a statement on the earnings call that encapsulates why the stock moved differently from Meta. “We are compute constrained in the near term. Our cloud revenue would have been higher if we were able to meet the demand.”
That sentence is the opposite of the problem Meta faces. Alphabet is not trying to convince investors that future revenue will eventually justify current spending.
It is telling them that current demand already exceeds current supply, and the spending is an attempt to keep up with paying customers.
When a company raising capex can point to a $460 billion backlog and say its revenue growth is being constrained by its inability to build fast enough, the market reads that capex as fulfilling proven demand.
When a company raising capex is still explaining that the returns will materialize in products that do not yet exist, the market discounts the uncertainty by selling the stock.
Google Cloud is now 18 percent of Alphabet’s total revenue. Its 63 percent growth rate more than doubled in a single quarter.
Pichai said the growth was led by enterprise AI solutions and AI infrastructure demand driven by adoption of Alphabet’s Gemini 3 model.
Ashkenazi said the company expects just over 50 percent of the $460 billion backlog to convert to revenue over the next 24 months, meaning Alphabet has visibility into approximately $230 billion in future Cloud revenue already booked.
That is the difference between a capex story that investors believe and one they are still waiting to see proven.
The Hyperscaler Picture
Meta and Alphabet reported alongside Microsoft and Amazon on Wednesday night, creating one of the most consequential single evenings of corporate earnings in the history of the technology industry.
Microsoft guided full-year 2026 capex to $190 billion, citing $25 billion attributable to higher component pricing, the same dynamic Meta and Alphabet cited. CEO Satya Nadella attributed the spending primarily to GPU and CPU demand for Azure and AI tools like Microsoft 365 Copilot.
Amazon held its capex budget at approximately $200 billion, the largest absolute spend of any of the four companies, while reporting 28 percent growth in AWS revenue.
Amazon’s stock fell despite the strong cloud numbers, as investors focused on the 95 percent year-over-year compression in free cash flow as infrastructure spending accelerated.
Combined, the four hyperscalers plus Apple are now on track to spend somewhere between $650 billion and $725 billion on AI infrastructure in 2026 alone, a figure that exceeds the GDP of most European countries.
The question that will define the next several years of technology investing is whether that spending produces commensurate returns across the economy or whether it represents the largest capital misallocation in corporate history.
The answers so far are mixed. Google Cloud’s 63 percent growth says the infrastructure is generating returns.
Amazon’s free cash flow compression says the bill has to be paid before the returns arrive.
Meta’s stock dropping 10 percent on a genuine earnings beat says the market is not yet sure which category Meta belongs in.
Matt Britzman, an analyst at Hargreaves Lansdown, put it simply. “The market was less united on what to make of the spending plans, with investors still trying to balance the scale of the AI opportunity against the cash required to chase it. But the bigger takeaway is that this cycle is nowhere near cooling.”
Meta’s Q2 revenue guidance came in at $58 billion to $61 billion, bracketing the analyst consensus of approximately $59.5 billion.
The advertising business continues to grow.
The AI investments that are not yet generating revenue are producing measurable improvements in the advertising products that are. The long-term bet may be exactly right.
The stock is down 10 percent today because the market wants more time before it is ready to agree.
