As the major AI cycle commences, the global technology industry is at the dawn of a structural transformation. Although market discussions continue regarding the pace of capital expenditure and returns, AI's potential to enhance productivity has firmly established its theme of long-term cyclical growth. The year 2026 is poised to be a critical observation period for monitoring AI infrastructure constraints, the evolution of the competitive landscape in chips, and the integration of end-user applications.
According to analysis from HSBC's Nicolas Cote-Colisson team in a report released on the 7th, AI is currently in the initial phase of a major cycle. Despite an instance in October 2025 where capital expenditure expectations outpaced initial revenue, considering the rapid pace of AI development and its potential to impact global GDP exceeding $110 trillion, the market maintains an optimistic outlook for 2026. The analysis suggests that risks, opportunities, and narrative focus will concentrate on six core dimensions.
On a macro level, HSBC holds an optimistic view of U.S. stock performance in 2026, setting a target of 7500 points for the S&P 500. Analysts advise investors to broaden their AI trade focus from pure infrastructure suppliers to include application users and enablers.
Global cloud giants Amazon.com, Microsoft, and Alphabet are experiencing significant order backlogs, yet capacity expansion is constrained by data center construction cycles. HSBC anticipates that capacity tightness will persist into 2026.
Against this backdrop, the willingness of tech giants to invest remains strong. Alphabet, Meta Platforms, Inc., and Amazon.com have all hinted at "significant/major" increases in capital expenditure for 2026. HSBC forecasts that the aggregate capital expenditure growth rate for relevant companies will reach 44% in 2026. With the exception of Oracle and CoreWeave, most giants are expected to fund these expenditures using free cash flow. HSBC believes the current spending pace is more limited by "construction capacity" than "willingness to invest," and that there is an upside risk to the 2026 expenditure guidance.
The most significant obstacle to capacity expansion is not funding, but power supply. According to HSBC analysis, the delivery cycle for heavy-duty gas turbines has extended to over four years, while small modular nuclear reactors (SMRs) are unlikely to contribute meaningfully to capacity before 2035. This implies that the power shortfall for data centers will persist in the short term.
In contrast, the Chinese market exhibits distinct characteristics. Domestic Chinese brands, benefiting from shorter delivery cycles, are rapidly penetrating the large engine market. Furthermore, liquid cooling technology will become a hot topic in 2026; as chip power increases, it will transition from an option to a necessity.
Alphabet, Meta Platforms, Inc., and Amazon.com have all signaled "significant/meaningful" capital expenditure growth for 2026, and HSBC expects most large tech companies to take similar action when issuing their 2026 guidance (late January/early February). HSBC projections indicate total annual capital expenditure will grow by 44%. Overall, funding is expected to come from current free cash flow generation, with the exception of Oracle and CoreWeave, which will require additional financing.
Given the aforementioned capacity constraints and robust demand, the report anticipates capital expenditure may opportunistically exceed initial guidance. Consequently, upside risk is expected for the 2026 guidance.
In the AI chip sector, NVIDIA's GPUs remain the preferred choice for cloud service providers, but competition from ASIC chips is intensifying. Alphabet's Ironwood TPU and Amazon.com's Trainium3 are representative examples. ASICs, with their superior performance-to-cost ratio and cost advantages, are attracting increased in-house development investment from hyperscalers.
HSBC expects the share of ASICs in cloud service providers' capital expenditure to rise from 2% in 2023 to 13% by 2027. Although external chip sales may not contribute substantially to financial statements until 2027, discussions surrounding orders and technology agreements in 2026 will influence valuations. NVIDIA's recent licensing agreement with Groq is viewed as a positive move to counter the trend of declining inference costs.
Competition in frontier large models is entering a consolidation phase. HSBC believes that high sunk costs will lead to market rationalization, ultimately resulting in an oligopolistic structure dominated by a few giants, alongside smaller specialized players. Currently, the intelligence level of open-source/open-weight models is rapidly catching up to proprietary models, with the gap narrowing to approximately three months.
Gemini's intelligence growth rate is surpassing that of ChatGPT. The focus for 2026 will be on monetization models for high-end models and the commercialization of AI advertising; HSBC forecasts that AI chat advertising will account for 2% of digital ad spend by 2030.
The year 2026 is seen as pivotal for the integration of AI into smartphones, while also marking the beginning of new hardware forms challenging traditional platforms. Apple is expected to significantly enhance the AI capabilities of its hardware in 2026-2027, including an upgraded Siri and potentially a foldable screen or a 20th-anniversary edition iPhone.
Smart glasses are considered a crucial medium for unleashing the power of large AI models. The collaboration between Meta Platforms, Inc. and EssilorLuxottica is currently leading the field, while Samsung, Google, Apple, and Amazon.com all have related products planned for launch around 2026-2027. These devices have the potential to partially supplant smartphone functions through voice interaction and contextual information capture.
