Key takeaways
- Microsoft generated $90 billion in Q4 FY26 revenue, driven by 43% growth in Azure, pushing the cloud business past $100 billion in annualized revenue.
- Meta missed EPS estimates at $6.18 (vs. $7.22 expected) as free cash flow collapsed 91% to $784 million due to massive AI infrastructure capex.
- Microsoft's commercial remaining performance obligations surged 84% to $678 billion, proving enterprise AI demand is translating into locked-in, multi-year contracts.
- The July 29 earnings split proves AI profits are currently landing at the cloud infrastructure layer, while model builders like Meta are absorbing the highest capital costs.
- Meta shares dropped roughly 10% on July 30, pushing the stock down nearly 30% year-to-date, signaling the end of Wall Street's grace period for unprofitable AI spending.
On July 29, 2026, Wall Street watched a tale of two tech earnings unfold in real time. Microsoft posted a staggering $90 billion in fiscal fourth-quarter revenue, sending its stock soaring nearly 15% in a single session and adding roughly $190 billion to its market cap. Hours earlier, Meta Platforms had watched its own shares plunge 10% after free cash flow collapsed by 91% under the crushing weight of artificial intelligence infrastructure spending.
The divergence paints a stark picture of the current AI boom. The capital expenditures required to build foundation models and train algorithms are massive, but the actual, monetized profits are currently landing squarely on the desks of cloud infrastructure providers. For retail investors holding Big Tech, the July 29 earnings cycle provided the clearest reality check yet: not all AI exposure is created equal, and betting on the wrong layer of the tech stack could bleed your portfolio dry.
Microsoft Cashes the AI Check
Microsoft didn't just beat Wall Street targets for the period ending June 30; it obliterated them. The company reported $90 billion in revenue (an 18% increase year-over-year) and adjusted earnings per share of $4.74. Net income jumped 31% to $35.77 billion, up from $27.23 billion a year prior. But the headline number was Azure.
The Intelligent Cloud segment was already a juggernaut, but Q4 confirmed that enterprise AI demand has transitioned from hype to signed contracts. Azure and other cloud services revenue grew 43% for the quarter, pushing the Azure business past the $100 billion annualized revenue mark for the first time in the company's history. Total Microsoft Cloud revenue hit $59.3 billion, up 27%.

The market's reaction was ferocious. According to Yahoo Finance, Microsoft (MSFT) soared 14.97% on the report, significantly outpacing the broader tech sector's gains. Reuters reported that the options market was bracing for a $190 billion market-value swing in either direction; the bullish outcome triggered a wave of short covering and institutional accumulation. Azure's 43% growth rate proved that Microsoft is successfully forcing enterprises to pay a premium for AI computing power, largely through its OpenAI partnership and proprietary Copilot integrations. When a company increases its commercial remaining performance obligations by 84% to $678 billion, as Microsoft disclosed, it means enterprise customers are locking themselves into multi-year, high-margin cloud ecosystems.
Meta Bleeds Cash in the Trenches
Contrast that with Meta's brutal Wednesday report. The social media giant actually grew its Q2 revenue by 28% to $60.8 billion, proving its core digital advertising engine remains highly lucrative. Ad revenue even rose 27% year-over-year. But the bottom line told a horror story.
Meta reported adjusted earnings per share of $6.18, badly missing the Street's consensus estimate of $7.22. Net income fell 14% year-over-year. According to company filings, Q2 expenses surged 55% to $42 billion, weighed down by massive AI infrastructure costs, a $2.4 billion legal contingency charge, and $1.2 billion in severance expenses. Operating income fell 8%.
The most alarming metric for shareholders was free cash flow. Reuters reported that free cash flow plummeted to just $784 million for the quarter, down from $8.55 billion in the same period a year ago. In Q1, the company had generated over $12 billion. Wall Street recoiled. According to Yahoo Finance, Meta shares plunged 10% in early Thursday trading, bringing the stock's year-to-date loss to nearly 30%. CNBC noted that CEO Mark Zuckerberg's aggressive buildout of AI data centers is swallowing almost all of the company's available cash.
The Great AI Bifurcation
The split between Microsoft and Meta isn't just a story of good and bad execution; it represents a fundamental bifurcation in the AI value chain. Microsoft functions as the digital landlord of the AI revolution. It owns the compute power, the data centers, and the enterprise software distribution channels. When companies want to deploy generative AI, they rent servers from Azure and buy licenses for Microsoft 365 Copilot. The revenue is high-margin, immediate, and massive.
Meta, conversely, is acting as the wildcatter. The company is spending unprecedented sums to train its Llama family of foundation models and upgrade its data center architecture to handle next-generation AI workloads. While Zuckerberg argues this investment is necessary to defend Meta's advertising dominance and build future products, the return on investment (ROI) is entirely speculative at this stage. Meta is spending tens of billions today for a theoretical payoff tomorrow. As of July 2026, AI is a proven cash machine for the cloud providers but an open financial wound for the model builders.

What the Q2 Numbers Mean for Your Portfolio
The immediate takeaway for investors navigating the August tech earnings season is that capital expenditure (capex) is the ultimate tiebreaker. When Microsoft spends on AI infrastructure, that spend is immediately validated by $678 billion in future contracted cloud revenue. When Meta spends, it simply shrinks the company's free cash flow by 91% and invites a 10% stock haircut.
This dynamic forces a reevaluation of the Magnificent Seven. Investors must begin separating infrastructure providers—Microsoft, Amazon (AMZN), and to a lesser extent Alphabet (GOOGL)—from consumer-facing platforms attempting to pivot their core businesses to AI. The market is no longer rewarding AI ambition; it is rewarding AI invoices. If a company cannot demonstrate that its AI capex is directly driving high-margin subscription or rental revenue, its stock is highly vulnerable.
Furthermore, the violent reaction to Meta's EPS miss underscores how low the market's tolerance has become for margin compression. For the last two years, investors gave tech giants a pass on free cash flow destruction under the guise of "winning the AI race." The Q2 2026 earnings cycle signals that the grace period is officially over. Profitability and cash generation matter again.
Actionable Steps for Tech Investors
If your portfolio is heavily skewed toward Big Tech, the events of July 29 demand a structural review.
- Audit your AI exposure by layer: Differentiate between the companies selling the AI "picks and shovels" (cloud compute, chips, data storage) and those building consumer-facing models. The infrastructure layer is currently capturing the lion's share of the profits. If you own the model builders, size those positions knowing they carry severe cash-flow volatility.
- Track commercial backlog, not just headline revenue: Microsoft's massive jump in commercial remaining performance obligations ($678 billion) was the hidden catalyst behind its 15% pop. When evaluating tech earnings, look for deferred revenue and contracted backlog to gauge the durability of future AI revenue.
- Monitor free cash flow over EPS: As Meta's quarter proved, an EPS miss can be triggered by one-time legal charges, but a 91% drop in free cash flow points to systemic capital drain. If a company's capex consistently outpaces its operating cash flow, it is financing AI through its balance sheet—a strategy that caps equity upside.
The article shows the pattern. The app trains the response.
Continue in Tikva to turn the insight into a repeated response.
Open TikvaSources and educational notice
This article is educational. It does not provide a medical diagnosis or replace guidance from a qualified health, legal, tax, investment, or financial professional. Decisions about your health or finances should consider your individual circumstances.
- CNBC — Microsoft Q4 FY26 earnings results, Azure revenue, and EPS beats
- Reuters — Meta's 91% drop in Q2 free cash flow and AI capex guidance
- CNBC — Meta's Q2 2026 EPS miss and dwindling free cash flow
- Microsoft Official Press Release — FY26 Q4 $90B revenue and Microsoft Cloud growth
- The Wall Street Journal — Azure surpassing $100 billion in fiscal year revenue and profit jumps
FAQ
Why did Microsoft stock surge while Meta stock dropped after July 29 earnings?
Microsoft jumped nearly 15% because it proved it is monetizing AI immediately, growing Azure cloud revenue by 43% and locking in $678 billion in future business contracts. Meta dropped roughly 10% because, despite growing ad revenue, its net income fell 14% and free cash flow collapsed by 91% as massive data center spending ate into profits.
What does the split between Microsoft and Meta mean for AI investing?
It highlights a major divide in the AI tech stack. The current profits are flowing to infrastructure and cloud providers like Microsoft that rent out computing power. Companies attempting to build foundational AI models, like Meta, are currently bearing the massive, unprofitable costs of that development without immediate, large-scale monetization.
How much did Meta spend on AI infrastructure in Q2 2026?
While exact quarterly AI capex varies, Meta's total Q2 expenses surged 55% to $42 billion, driving free cash flow down to just $784 million from $8.55 billion a year prior. CEO Mark Zuckerberg has committed to spending between $125 billion and $145 billion on AI and data center infrastructure, a cost Wall Street is currently penalizing the stock for.