Four Tech Giants Bet $725 Billion on AI

Four Tech Giants' AI Capital Expenditure Soars to $725 Billion

On April 30, the Financial Times compiled first-quarter earnings data from major tech giants, revealing a market-shaking figure — Google, Amazon, Microsoft, and Meta plan to invest a combined $725 billion (approximately 4.96 trillion yuan) in capital expenditure in 2026, a 77% increase over last year's record-setting $410 billion. The data once again confirms that the world's top tech companies are betting on AI infrastructure with unprecedented intensity.

Facing a steady stream of AI bubble narratives in the market, Jefferies analyst Brent Thill pushed back without hesitation: "The AI economy is healthy, and recent revenue growth is more than enough to support these massive investments. The bear case is garbage."

Google Leads the Pack: Cloud Revenue Surges 63% Year-over-Year

Among the four giants, Google's performance has been particularly impressive. Its cloud business revenue grew 63% year-over-year to $20 billion, making it the most prominent benchmark case for AI investment returns. This growth not only far exceeded market expectations but also provided solid financial backing for Google's continued push into AI infrastructure.

Google's strong performance indicates that demand for enterprise AI services is accelerating, and cloud computing — as the core delivery vehicle for AI capabilities — is entering a new phase of explosive growth.

Microsoft: $190 Billion Spend Still Not Enough to Meet Demand

Microsoft set its calendar year 2026 capital expenditure at $190 billion, significantly above the analyst consensus estimate of $152 billion — a 25% overshoot. Microsoft CFO Amy Hood revealed that approximately $25 billion of the increase was driven by rising prices for memory chips and components.

More notably, Hood candidly told investors that even with the substantial spending increase, Microsoft will remain constrained by capacity bottlenecks through at least 2026. The company is accelerating the deployment of GPU, CPU, and storage infrastructure, but the pace of AI compute demand growth continues to outstrip supply expansion. This statement implies that the supply-demand gap in AI infrastructure will be difficult to close in the near term.

Meta: Budget Raised by $10 Billion in All-In AI Bet

Meta also significantly ramped up its investment, raising its full-year capital expenditure forecast by $10 billion, with the upper end of the range climbing to $145 billion. Meta attributed the increase to rising component prices, particularly the sustained surge in memory chip costs.

Rising memory chip prices have become a common factor driving up spending across multiple tech giants. As AI model parameter counts continue to expand and training data volumes grow, demand for core components such as High Bandwidth Memory (HBM) has surged, further intensifying supply chain tensions.

The Deeper Logic: AI Investment's 'Flywheel Effect' Has Kicked In

A key signal can be distilled from the earnings data of all four tech giants: AI investment is forming a virtuous cycle. Exemplified by Google Cloud's 63% revenue growth, the commercial returns from AI services are materializing rapidly, and these returns in turn provide the confidence for even larger infrastructure investments.

This logic chain also explains why analysts are dismissive of AI bear cases. Unlike the cash-burning model of the dot-com bubble era, which lacked revenue support, current AI investments are anchored by clearly visible revenue growth.

Of course, risks are not absent. The staggering $725 billion in expenditure means that any slowdown in AI demand growth or underperformance in commercialization timelines could place enormous financial pressure on these companies. Additionally, sustained increases in chip and component prices are eroding investment efficiency.

Outlook: The AI Arms Race Is Far from Peaking

Based on current trends, the tech giants' AI arms race is still accelerating. Microsoft's acknowledgment of capacity constraints suggests that further increases in capital expenditure are possible. Meanwhile, upstream chip manufacturers such as NVIDIA and AMD stand as direct beneficiaries of this investment wave.

For the AI industry as a whole, the nearly 5 trillion yuan in capital inflows will significantly accelerate the expansion of computing infrastructure, providing stronger foundational support for large model training, inference deployment, and the rollout of various AI applications. The AI industry's "infrastructure era" is unfolding at a pace that exceeds expectations, and the endgame of this race is perhaps still far from sight.