Markets

Microsoft Cloud Thrives as Meta Sinks on AI Costs

  • July 30, 2026
  • 6 min read
Microsoft Cloud Thrives as Meta Sinks on AI Costs

On Wednesday, two of the world’s largest technology companies reported their quarterly financial results. Both Microsoft and Meta Platforms are spending historic amounts of capital to build out artificial intelligence infrastructure.

Wall Street delivered two entirely different verdicts.

Microsoft shares climbed in after-hours trading, gaining roughly 8 percent immediately following the report before settling higher amid a broader market selloff. Investors rewarded the software maker for showing clear, direct returns on its AI spending through its cloud computing division.

Meta’s stock fell up to 10 percent after the closing bell. Despite beating revenue expectations, the social media parent reported a 14 percent drop in net income and a severe contraction in free cash flow, raising concerns about how long it will take to monetize its massive server and data center investments.

The divergence marks a shift in how financial markets treat the ongoing artificial intelligence boom. For the past two years, investors have largely rewarded companies simply for announcing AI initiatives. Now, the market is demanding proof that the tens of billions of dollars being poured into graphics processing units (GPUs), data centers, and power grids will actually generate a financial return.

Microsoft’s Cloud Business Proves the Model

Microsoft reported fiscal fourth-quarter revenue of $90.01 billion, an 18 percent increase from the same period last year, easily beating analyst expectations of $87.62 billion. Net income rose 31 percent to $35.8 billion.

The primary driver was Azure, Microsoft’s cloud computing platform. Azure revenue grew by 43 percent in the quarter. Across the full fiscal year of 2026, Azure crossed $100 billion in annual revenue for the first time, officially making it larger than Google Cloud.

Microsoft spent $41 billion in capital expenditure during the quarter, largely to expand its AI infrastructure. That figure is up more than 60 percent from a year earlier. However, the company is already turning those capital expenses into recurring revenue by renting computing power to other businesses and selling software subscriptions. Microsoft 365 Copilot, the company’s AI assistant, passed 30 million paid users during the quarter.

Charles Lamanna, Microsoft’s executive vice president for Copilot, Agents and Platform, told reporters the company has reached a tipping point with corporate customers expanding from small trials involving a few hundred staff to broad deployments covering tens of thousands of employees.

“Cloud and AI is the driving force of business transformation across every industry and sector,” Microsoft Chairman and CEO Satya Nadella said in the earnings release.

While Microsoft’s free cash flow did fall 23 percent year-over-year, it remained healthy at $19.64 billion. This provided the market enough comfort to tolerate the company maintaining its aggressive full-year capital expenditure outlook of $175 billion.

Meta Strains Under Infrastructure Costs

Meta reported a different financial reality. The company’s core advertising business performed well, with second-quarter revenue rising 28 percent to $60.8 billion, surpassing forecasts of $60.17 billion.

But the cost of running the business surged. Total expenses jumped 55 percent year-over-year to $42.03 billion. Net income fell 14 percent to $15.8 billion, and earnings per share came in at $6.18, missing Wall Street expectations of $7.22.

The steepest drop appeared in free cash flow, which collapsed by 91 percent, falling from $8.55 billion a year earlier to just $784 million. Meta is currently spending cash faster than its advertising business brings it in.

During the quarter, Meta spent about $31 billion on capital projects. The company also raised the lower end of its capital expenditure forecast for 2026, telling investors it now expects to spend between $130 billion and $145 billion this year. Just last year, Meta’s capital spending was $72 billion.

The expense sheet included more than just servers. Meta booked a $2.4 billion charge tied to legal proceedings and $1.18 billion in severance costs following May layoffs that impacted roughly 8,000 employees. Meanwhile, Reality Labs, the division responsible for virtual reality headsets and wearable devices, reported an operating loss of $4.6 billion on just $431 million in sales.

Meta CEO Mark Zuckerberg defended the spending, citing strong momentum in the company’s AI research and product integration. He argued that these investments are actively improving recommendation algorithms and creating new tools for advertisers. The company reported that its family of apps—Facebook, Instagram, WhatsApp, and Messenger—now reaches 3.6 billion daily active people.

The Structural Difference

The contrasting earnings highlight a fundamental difference in how Microsoft and Meta operate.

When Microsoft or Alphabet buys specialized AI chips from Nvidia, they place them in data centers and rent the computing capacity directly to other developers and enterprises. The spending translates directly to a recognized revenue line in their cloud divisions.

Meta does not have a commercial cloud business. It purchases the same expensive hardware to train its own models, like Llama, and to improve user engagement on Facebook and Instagram. While better algorithms eventually lead to higher ad prices—Meta reported ad impressions rose 14 percent and prices increased 12 percent—the financial return is indirect and takes longer to materialize.

To help manage the sheer scale of the costs, Meta is beginning to move some infrastructure financing off its balance sheet. Earlier this week, the company announced a $14 billion data-center joint venture in Texas with BlackRock, which will own 80 percent of the campus.

Accounting Tactics Mask Deeper Costs

Financial analysts have pointed out that Meta’s actual infrastructure costs are taking a heavier toll on the business than its headline earnings suggest.

When technology companies purchase expensive servers, they do not book the full cost immediately. They spread the cost over the expected lifespan of the equipment through depreciation.

Recently, Meta extended the estimated useful life of its server hardware to 5.5 years. This accounting adjustment spreads the depreciation over a longer period, reducing the expense recorded in any single quarter and artificially boosting reported profits.

If Meta had maintained a shorter, three-year depreciation schedule—which aligns more closely with the actual product cycle of advanced AI chips—analysts estimate the company’s reported earnings per share would be roughly 15 percent lower. With planned AI capital expenditures pushing $145 billion this year, any future realization that these chips become obsolete faster than 5.5 years could force Meta to recognize massive depreciation hits, further damaging profit margins.

Microsoft and Alphabet use similar accounting methods, but their absolute profit numbers are large enough to absorb potential depreciation shocks without breaking their balance sheets. Meta’s shrinking free cash flow leaves it with far less margin for error.

What Happens Next

The technology sector is entering a more selective phase. The broad market selloff leading into Wednesday’s earnings—driven partly by a U.S. Federal Reserve meeting and surging long-end Treasury yields—showed that investors are losing patience with companies that promise future AI profits without delivering current cash flow.

For Microsoft, the path is clear. As long as Azure continues growing at a 40-plus percent clip, Wall Street will likely tolerate the company’s $175 billion infrastructure build-out.

For Meta, the pressure is mounting. The company must prove to a skeptical market that its unprecedented spending will result in measurable financial growth before its cash reserves deplete further.

About Author

Jennifer Gross

Jennifer Gross is a technology and business writer with a passion for covering emerging innovations, digital trends, startups, AI, cybersecurity, and the future of online business. She specializes in breaking down complex tech topics into practical, engaging insights for everyday readers and industry professionals alike. Through her work with Tech Journal HQ, Jennifer explores the evolving intersection of technology, entrepreneurship, and modern digital culture.