Hexun Investment Advisor Wang Jianghong: The National Destiny Battle of AI Between China and the US

Friends, who will ultimately win the race for national destiny in AI between China and the United States? Today, using Nvidia founder Jensen Huang’s “Five Layers of AI Cake” theory, we analyze why this competition will ultimately favor China.

The AI industry is like a five-layer cake, from bottom to top: energy, chips, infrastructure, models, and applications. The bottom layer determines the ceiling; whoever solves the bottleneck at this level first will be the last to laugh.

Let’s first look at the respective weaknesses of China and the US. America’s Achilles’ heel is energy, while China’s shortcoming is chips. According to authoritative data in 2025, China’s total power generation reaches 10.58 trillion kWh, while the US has only 4.76 trillion kWh. China’s power generation is 2.22 times that of the US. More critically, 70% of US power grid equipment is aging, and cross-state transmission approval cycles can take over a decade, leading to severe power shortages for data centers. Even if US chips are powerful, without electricity support, they are just scrap metal. This is a dual hardware problem at the institutional and physical levels, difficult to resolve in ten years.

In contrast, China’s chip weakness is a technical issue, which can be overcome. Huawei has already stated that there are no issues with chips. Why can they say that? Because we have a complete industry chain, a large team of engineers, and the advantages of a national coordinated system. Domestic AI chips have already achieved breakthroughs from zero to one. Relying on stacking technology and algorithm optimization, they can support mainstream training and inference needs, without hindering China’s booming AI industry. Comparing the two, the US’s power bottleneck is a dual hardware problem at the physical and institutional levels, while China’s chip bottleneck is a stage-specific technical and engineering challenge. Overcoming China’s shortcoming is far easier than resolving the US’s energy issues.

Looking at the application layer, China’s advantages are even more overwhelming. The value of AI ultimately depends on its application. China has the world’s largest single digital market and the most complete industrial categories. Over 1.1 billion internet users provide vast application scenarios and rich data loops for AI. In manufacturing, e-commerce, logistics, and other fields, China’s AI penetration rate is already far ahead. For example, the US industrial AI penetration is about 67%, while AI application in China’s e-commerce sector exceeds 80%. This data-driven feedback loop of algorithms and scenarios creates a perfect positive cycle of data, algorithms, and applications. China’s policies are highly coordinated, infrastructure is well-developed, and ultra-high-voltage power grids ensure low-cost computing power. Intelligent computing centers are rapidly expanding nationwide, and the adoption of lightweight AI applications among small and medium-sized enterprises is leading globally. Although the US focuses on technological innovation and high-end scenarios, market fragmentation, high implementation costs, and slow commercialization pace make it impossible to compete with China’s scale advantages.

In summary, the US is trapped by energy constraints, unable to move freely, while overcoming China’s chip bottleneck is only a matter of time—and that time will not be long. China’s market size and AI application ecosystem have already established an absolute lead, exchanging scale for efficiency and driving innovation through deployment. In the medium to long term, the race for AI national destiny between China and the US will see China achieve a counterattack, maintain its lead, and ultimately win—an inevitable trend.

(Edited by: Zhang Yan)

【Disclaimer】This article only reflects the author’s personal views and has no relation to Hexun.com. Hexun.com remains neutral regarding the statements and opinions expressed in this article and does not provide any explicit or implicit guarantees regarding the accuracy, reliability, or completeness of the content. Readers are advised to use this for reference only and bear all responsibilities themselves. Email: news_center@staff.hexun.com