Microsoft and Meta Stock Prices Diverge, Perfectly Illustrating How Markets Evaluate AI Investments

Multiple foreign media outlets reported on Thursday (30th) that Microsoft (MSFT-US) and Meta (META-US) have both invested over $100 billion in expanding AI data centers, yet received dramatically different market evaluations in their latest financial reports. The key is not whether the two companies benefit from AI, but whether massive capital expenditures can be quickly converted into revenue, profit, and free cash flow.

Microsoft's stock surged 15.51% on Thursday, marking its best single-day performance since 2008; Meta, however, fell nearly 8%, closing lower for the 11th consecutive trading day—the longest losing streak since its IPO—and cumulative losses during this period exceeded 20%.

Both companies maintained strong revenue growth in their latest quarter, but Microsoft has established a relatively clear AI monetization model through its Azure cloud services and Copilot software; Meta's AI infrastructure is currently used primarily internally. While investment efficiency is reflected in its advertising business, it has not yet formed an independent revenue stream that can be directly charged.

Microsoft's latest quarterly revenue exceeded market expectations, with Azure revenue growing 43%, surpassing analysts' forecasts. Paid seats for Microsoft 365 Copilot surpassed 30 million, up from over 20 million announced in April, indicating that enterprise customers are accelerating their adoption of AI tools.

Microsoft's enterprise software division has exceeded expectations for 16 consecutive quarters, and profitability has improved compared to the same period last year, alleviating market concerns that generative AI might disrupt traditional software subscription models. Although the Intelligent Cloud segment, which includes Azure, faces pressure from rising depreciation expenses, it has maintained its operating margin at the same level as the previous year through operational efficiency improvements.

Cash flow performance has widened the gap between the two companies.

Microsoft generated nearly $20 billion in cash flow after deducting capital expenditures in the quarter, sufficient to fund share buybacks and future investments. The company also maintained its capital expenditure outlook for fiscal 2026 unchanged, and although it hinted at possible further expansion in fiscal 2027, investors believe that the growth of Azure and Copilot has already demonstrated that these investments are beginning to yield returns.

In contrast, Meta's core advertising business is also strong. The company's second-quarter revenue increased 28% year-on-year, the best performance since the pandemic-driven high-speed growth in 2021; ad impressions increased 14%, and average ad prices rose 12%, reflecting that AI recommendation and ad placement tools are improving platform efficiency.

However, strong revenue growth failed to offset the pressure from expenses. Meta's quarterly free cash flow declined 91% year-on-year to $784 million, leaving 'almost no cash' after capital expenditures, and the company has suspended share buybacks for three consecutive quarters.

Meta's R&D expenses surged 67% year-on-year, accounting for 36% of quarterly revenue, while its operating margin declined 12 percentage points year-on-year, and earnings per share were significantly below market expectations.

The company expects this quarter's revenue to be between $61 billion and $64 billion, with a midpoint of $62.5 billion, still below analysts' forecast of $63.15 billion.

The fundamental difference between the two companies lies in the use of AI computing resources.

Microsoft can rent out data center capacity to external customers via Azure, while also charging software subscription fees through Copilot, allowing capital expenditures to be more directly converted into revenue.

Meta's newly built AI servers are primarily used for internal model training, inference, and product development, lacking a mature external monetization platform like Azure.

Meta CEO Zuckerberg stated that the company has received numerous offers to lease computing resources, with some parties even willing to pay prices higher than Meta's setup costs, hinting at the possibility of developing cloud services in the future.

However, CFO Susan Li later stated that the company is currently prioritizing the use of computing power for its own training and inference needs, and the specific model and timeline for external leasing remain unclear.

The divergent stock performance of Microsoft and Meta on Thursday highlights that the market is not opposed to tech giants expanding AI investments, but is now demanding more concrete financial returns.

Microsoft has proven the effectiveness of its investments through Azure growth, Copilot paid users, and abundant cash flow; Meta, while successfully using AI to boost ad revenue, faces a significantly smaller margin for error in its AI strategy amid continuously increasing capital expenditures, rapidly shrinking free cash flow, and the absence of new revenue sources.