Goldman Sachs Recommends Long-Term Exposure to China’s AI Value Chain Amid Low Global Fund Allocation

• Goldman Sachs recommends buying a basket of China’s AI value chain, covering power, semiconductors, AI infrastructure, models, and applications.
• Goldman Sachs estimates that China's AI-related market capitalization is approximately $4 trillion, accounting for about 16% of global AI-related revenue, yet allocations to China in global mutual funds' technology exposure remain at only about 1.2%.
• The core of this transaction is not the surge of a single AI application, but the revaluation opportunity driven by undercapitalized funding, policy investments, and hardware demand.
• The risk lies in the continued realization of data center investments, storage capacity expansion, IPO financing, and AI hardware exports.

Goldman Sachs' Thematic Research team is bringing the "China AI Value Chain" to the center of trading attention.

According to its report titled "Trading Strategy: Long China’s AI Value Chain," Goldman Sachs recommends going long on a China AI basket covering power, semiconductors, AI infrastructure, models, and applications. Over the past two years, global AI trading has been dominated by U.S. big tech stocks, NVIDIA’s supply chain, and cloud capital expenditures; Goldman Sachs now sees a misalignment between China’s AI assets in terms of market capitalization, revenue contribution, and global capital allocation.

According to Goldman Sachs' estimates, Chinese AI-related companies have a combined market capitalization of approximately $4 trillion and contribute about 16% of global AI-related revenue, yet as of January 2026, global mutual fund managers allocate only about 1.2% of their global technology exposure to China.

This set of numbers forms the core trading logic of the entire report: if China’s AI industry already holds a double-digit share of global revenue, yet global capital allocation remains significantly underserved, then there is potential for the Chinese AI value chain to be repriced.

Goldman Sachs provided a straightforward comparison of global AI assets.

Since the end of 2022, global AI-related stocks have generated approximately $34 trillion in market capitalization, with China’s AI-related market capitalization reaching about $4 trillion, accounting for roughly 10% of the global total. In terms of revenue, China contributes approximately 16% of global AI-related income.

The allocation of funds is far below this percentage. Goldman Sachs estimates that, as of January 2026, global mutual fund managers allocated only about 1.2% of their global technology exposure to China.

This is also the core reason why Goldman Sachs has proposed going long on China’s AI value chain. U.S. AI assets have been repeatedly bought by global capital, with NVIDIA, cloud providers, semiconductor equipment, and power infrastructure all incorporated into the core AI trading theme. In contrast, although Chinese AI assets have already achieved a certain scale of revenue, they remain underweighted in global fund portfolios.

In other words, Goldman Sachs is not betting on the broad "China AI narrative," but rather on a more specific gap in capital allocation: revenue contributions have already materialized, but global portfolio holdings have not yet caught up.

Goldman Sachs emphasized that this transaction differs from traditional KWEB trades.

KWEB typically represents exposure to China’s internet and platform economy, prompting investors to think of e-commerce, advertising, online entertainment, and local services. However, Goldman Sachs has constructed the GS China AI Value Chain (GSXACART) basket, which spans from power, semiconductors, and AI infrastructure to models and applications—more closely aligning with a comprehensive Chinese AI supply chain.

Under this framework, the location of hardware and infrastructure comes first.

China’s push for technological self-reliance and the development of advanced computing capabilities has drawn simultaneous attention from policy makers, industry, and capital to AI hardware, data centers, power infrastructure, and semiconductors. Goldman Sachs believes that the value of these sectors has not yet been fully reflected in the stock market.

Their research estimates that the potential economic benefits from AI—driven by efficiency gains and the creation of new profits—could be 50% to 100% higher than what is already reflected in current AI stock prices. This is also why power, AI infrastructure, and semiconductors are positioned at the core of the basket.

Whether models and applications can explode in growth ultimately depends on computing power, storage, electricity, and equipment supply—all areas where China excels in large-scale manufacturing, engineering construction, and industrial ecosystem support.

Changes in China's AI hardware supply chain are moving from concepts to more concrete orders, exports, and financing milestones.

On the demand side, customs data cited by multiple media outlets showed that China’s exports rose 19.4% year-over-year in May, marking the strongest growth in three months. Among these, export value of integrated circuits surged approximately 111%, while export volumes increased only slightly. Price and structural shifts suggest that demand for AI hardware is one of the key driving factors. For memory, semiconductor equipment, and upstream materials, such data points to improved order volumes and higher capacity utilization.

On the policy and investment front, according to Reuters citing Bloomberg, China is preparing a five-year plan worth approximately RMB 2 trillion, or about $295 billion, to build a nationwide network of AI data centers. The plan has not yet been officially announced, but if implemented, it will directly drive demand for domestic storage chips, semiconductor equipment, power infrastructure, and data center facilities.

On the capital markets side, public reports indicate that A-shares, Hong Kong stocks, and certain global indices will increase the weightings of AI and semiconductors in their 2026 adjustments. This will enhance the visibility of passive capital flows to related companies and attract more domestic and international capital toward advanced computing and semiconductors.

Individual companies and industry cases further reinforce this trend. Yangtze Memory Technologies reported a year-over-year revenue increase of approximately 445% in the first quarter of 2026, with its global NAND flash market share rising from 8% to 13% over the past year, placing it in a tie for fourth place, while advancing its domestic IPO plan to support capacity expansion.

ChangXin Memory Technologies is regarded as a key player in China's DRAM industry. Third-party research estimates its 2026 revenue could exceed $50 billion; according to its prospectus, the company reported revenue of RMB 50.8 billion in the first quarter, with a revenue guidance of RMB 110 to 120 billion for the first half of the year.

These cases do not mean that Chinese storage companies have fully caught up with overseas giants, but they indicate that China’s AI hardware supply chain is transitioning from a “policy concept” to more tangible milestones in revenue, market share, funding, and production expansion.

Goldman Sachs also noted that China’s AI sector has outperformed other China-related assets and shown signs of capital reallocation. However, China’s AI assets still significantly lag behind their U.S. counterparts.

This is also where trading appeal and risk boundaries coexist.

The appeal lies in the fact that if global investors continue to seek growth opportunities beyond U.S. AI, the underweight positioning of Chinese AI could create room for capital reallocation. Especially after valuations of leading U.S. AI companies have already risen significantly and capital expenditure expectations have been thoroughly discussed, the market will naturally look for supply chain and application assets that have not yet been fully held.

The risk is that this remains a trading suggestion, not an established industry conclusion. The 2 trillion RMB AI data center initiative depends on policy details and actual implementation; listing, capacity expansion, and profitability improvements by companies such as CXMT and YMTC will also take time; whether chip export and sales data can be sustained will depend on the global AI hardware cycle and trade environment.

U.S. AI remains the primary global benchmark for capital allocation. Whether in model capabilities, cloud provider capital expenditures, GPU ecosystems, or enterprise application revenues, the U.S. market still offers a more mature standard. For Chinese AI to attract more global capital, it must do more than prove it is "undervalued and underweighted"—it must consistently deliver revenue growth, profitability, and technological advancements.

The key takeaway from Goldman Sachs’ bullish stance on China’s AI value chain is not that China’s AI has caught up to the U.S., but rather that it highlights a market misalignment: China accounts for approximately $4 trillion in market capitalization and about 16% of global revenue contribution, yet represents only around 1.2% of global mutual funds’ technology exposure.

Whether the funding can cover this gap depends on whether policy investments, hardware demand, and corporate profitability can continue to materialize.