Financial data indicates that overseas cloud service providers (CSPs) are maintaining a high growth trajectory in capital expenditure. In the first half of the 2026 calendar year, total capital expenditure for the four major CSPs reached $292.7 billion, a year-on-year increase of 83.64%. The 2026 guidance range of $720 billion to $745 billion has been raised from the initial $695 billion to $725 billion earlier this year.
The capital expenditure achievement rate for the four major CSPs in the first half was approximately 39.29% to 40.65% of the full-year guidance. This implies that capital expenditure in the second half of 2026 is expected to be 45.99% to 54.52% higher than in the first half, reflecting a long-term commitment to infrastructure development and strong optimism about the AI industry's future. The high growth in cloud services and a growing backlog of orders are validating the accelerating commercialization of AI, with capital expenditure increasingly converting into predictable revenue.
Cloud Growth and Order Backlog Confirm AI Commercialization Acceleration
Microsoft's Azure and other cloud services revenue grew 43% year-on-year, with annualized Azure revenue surpassing $100 billion for the first time. Its commercial remaining performance obligations reached $678 billion, an 84% increase, with new orders primarily coming from traditional enterprise customers outside of OpenAI, indicating a more diversified demand structure. Amazon Web Services (AWS) revenue grew 37% to $42.2 billion, its fastest pace in 18 quarters, with AI-related annualized revenue exceeding $25 billion and a backlog of $496 billion. Google Cloud revenue increased 82% to $24.77 billion, with its backlog exceeding $514 billion, the number of new customers doubling year-on-year, and actual usage by existing customers exceeding initial commitments by 50%. Cloud business operating margins improved to 35.6%. Meta's cloud-related revenue grew 27.96% to $60.801 billion, the fastest among the four CSPs, and its AI-driven Advantage+ advertising products now have an annual revenue run rate exceeding $75 billion.
The high growth in cloud revenue and the substantial order backlog for CSPs suggest that AI is moving from proof-of-concept to large-scale monetization. Capital expenditure is being efficiently converted into reliable revenue, providing the internal momentum for continued investment.
Financial Health Supports Long-Term Investment Despite Diverging Cash Flow Metrics
Microsoft's Azure AI service gross margin exceeds 70%, and the company has committed to maintaining positive free cash flow through fiscal year 2027. Amazon's AWS operating margin improved to 39.4%, and the high margins of its AI business are expected to gradually improve cash flow. While Google's free cash flow turned negative at -$5.855 billion in the second quarter of 2026, its cloud business operating margin rose to 35.6%, suggesting that economies of scale will help alleviate cash flow pressure. Although Meta's AI investments are entirely funded by advertising revenue from its own ecosystem, its AI-driven advertising tools have achieved an annualized revenue run rate of $75 billion, demonstrating significant long-term monetization potential. The leading CSPs are balancing investment and returns by optimizing capital structures and improving operational efficiency, with the high growth and high margins of their AI businesses providing strong financial support for long-term spending.
Risks
Risks include the potential for AI development to fall short of expectations and the possibility that application deployment may not proceed as anticipated.
Comments