Tech giants are ramping up AI infrastructure investments, with "burning cash" becoming the dominant theme in the U.S. tech sector.

Throughout July, the "Magnificent Seven" U.S. tech stocks, AI computing, and semiconductor companies displayed a highly unified yet fragmented market trend. The sector saw a collective pullback early in the month, with the Philadelphia Semiconductor Index plunging 11% in a single week. Nearly 70% of semiconductor stocks retreated over 20% from their highs, as investors worried that massive capital expenditures would continue to erode corporate cash flows.

The tide turned dramatically after the concentrated earnings reports at month-end. Microsoft surged 15.51% in a single day, its biggest daily gain since 2008, directly triggering a rally across memory, equipment, and AI chip stocks. The Philadelphia Semiconductor Index soared 8.19% in a single day, with Micron, SanDisk, and AMD all posting gains exceeding 13%.

All major companies shared a common strategic focus in July: continuously raising full-year capital expenditure guidance, with funds overwhelmingly directed toward AI servers, data centers, self-developed chips, and memory capacity. According to a rough estimate by Jiemian News, major tech companies including Amazon, Alphabet, Microsoft, and Meta collectively disclosed nearly $800 billion in full-year capital expenditure plans in a single month.

Google's parent company, Alphabet, was the first to raise its full-year capital expenditure guidance in mid-July to $195 billion-$205 billion. Alphabet CFO Anat Ashkenazi stated on the earnings call that Google Cloud's backlog exceeds $514 billion, with a computing capacity gap expected to persist for at least two years.

Microsoft maintained its original $190 billion capital expenditure plan but reported quarterly capital spending of $41 billion, a 70% year-over-year increase, all allocated to purchasing GPUs to expand the Azure computing cluster. The company's paid Copilot user base continues to grow rapidly, with investments translating into revenue, which was the core driver behind Microsoft's stock surge at month-end.

Meta and Amazon also signaled increased spending at the end of July. After releasing its second-quarter earnings, Meta raised its full-year capital expenditure guidance from $125 billion to $130 billion-$145 billion, with quarterly capital spending reaching $31.08 billion, a significant year-over-year increase.

Meta CFO Susan Li stated on the call that the industry faces a long-term shortage of computing capacity, and the company will maximize data center construction speed between 2026 and 2027, while advancing a $14 billion supercomputing campus project in Texas and a $50 billion project in Louisiana.

Unlike other cloud giants, Meta does not have a cloud business to sell computing capacity externally, meaning all AI investments must be absorbed through ad revenue. The company's free cash flow plummeted 91% year-over-year to $784 million in the second quarter, leading to an 8% stock price decline in after-hours trading following the earnings release, making it a weak performer in the July capital expenditure expansion rally.

Amazon also announced a major expansion plan. CEO Andy Jassy stated on the earnings call on July 30 that the company's full-year capital expenditure guidance was raised again to $220 billion, an increase of $20 billion from the initial forecast in February.

Jassy predicted that the supply-demand imbalance in computing capacity would persist until 2028. "Even at this level of spending, we still cannot meet all computing capacity needs in 2026, and demand will remain strong in 2027 and 2028," he said.

Amazon Web Services (AWS) reported second-quarter cloud revenue of $42.2 billion, a 37% year-over-year increase, the highest in 18 quarters. AI business annual revenue exceeded $25 billion, achieving triple-digit growth. Robust cash flow offset the pressure from capital expenditures, and Amazon's stock price surged nearly 10% in after-hours trading following the earnings release, a stark contrast to Meta's performance.

Hardware giant Apple also released its latest earnings report on July 30. For the third fiscal quarter of fiscal 2026, the company reported broad-based growth in hardware revenue, but weak guidance for the fourth quarter and slower-than-expected growth in services revenue led to a 6.33% decline in after-hours trading.

Although Apple has not built large-scale AI data centers, it continues to increase investment in the research and development of on-device AI chips while securing long-term supply agreements for memory and advanced chips upstream. This is essentially an extension of the cost expansion logic in the AI industry chain, complementing the computing capacity expansion of other tech giants.

Additionally, Tesla reiterated in July that its full-year capital expenditure would exceed $25 billion. The company's self-built AI chip factory in Texas is now operational, primarily relying on on-device AI for autonomous driving to achieve computing capacity expansion.

As direct beneficiaries of capital expenditure, semiconductor companies' operations in July were fully aligned with the tech giants, forming a supply-demand closed loop.

Intel released its earnings report on July 24, with quarterly revenue of $16.1 billion, a 25% year-over-year increase. Data center AI revenue surged 59% year-over-year, and the company raised its full-year capital expenditure guidance to $20 billion. Intel is expanding its advanced process factories in the U.S. and has signed ten long-term server CPU supply agreements with cloud providers to lock in orders.

Nvidia's stock price fluctuated downward in July, hitting a low of $190 during the session. Investors worried that continued expansion by cloud providers could lead to an oversupply of chips. However, strong end-user demand data, coupled with a market rebound at month-end, helped the stock close 2.65% higher, maintaining a market capitalization of $4.7 trillion.

AMD, memory manufacturers Micron and Western Digital, and equipment maker Applied Materials all secured large-scale long-term orders from hyperscale cloud providers in July. Memory chip companies benefited from increased AI computing storage demand, leading to a collective surge at month-end, with SanDisk posting a single-day gain of nearly 26%.

Regarding the current round of computing capacity expansion, some analysts suggest that capital is rotating from the "Magnificent Seven" to the semiconductor sector. The market is rewarding beneficiaries of capital expenditure while questioning cloud providers that continue to burn cash, indicating that the short-term divergence in market trends will persist.

BlackRock, the world's largest asset manager, believes the recent sell-off is an "overreaction," arguing that the market is conflating "a shift in the AI competitive landscape" with "a collapse in AI investment." BlackRock emphasized that cheaper AI will not end investment but will instead accelerate the adoption of AI applications across industries, expanding the potential market size and driving up overall demand for infrastructure such as data centers, memory chips, and electricity. The firm maintains an overweight position on U.S. stocks and recommends focusing on AI bottleneck areas such as chips and power.