Four years ago the United States tightened export controls on advanced semiconductors used in AI, data centers and defense, aiming to limit Beijing’s ability to boost its military and economic power. Those restrictions accelerated China’s long‑standing push for chip self‑reliance, a goal from the Made in China 2025 program, and prompted Beijing to pour hundreds of billions of dollars into domestic semiconductor production.
Beijing has offered large subsidies, tax breaks and other incentives to nurture domestic rivals to US and Taiwanese leaders such as NVIDIA and TSMC. The mainland foundry SMIC, central to China’s strategy, posted record revenues of $9.3 billion last year, while HuaHong reported operating above capacity amid surging demand.
Despite rapid investment and scale‑up, outside experts say China remains short of full self‑sufficiency. Ryu Yongwook of the National University of Singapore notes China lags the US in research, design and innovation and trails Taiwan and South Korea in advanced production.
Still, China has achieved significant gains. According to the Rhodium Group, Chinese firms now hold about 30% of the global market for legacy chips — widely used in vehicles, industrial equipment and consumer electronics. Mass production of these less advanced but essential semiconductors has driven down global prices and put pressure on non‑Chinese vendors, especially in areas like silicon carbide wafers used for high‑power chips.
On more advanced nodes, China has produced 7‑nanometer‑class processors powering Huawei’s latest smartphones — comparable to TSMC chips from 2018 but behind 3nm and 5nm designs in speed, power efficiency and cost. Analysts warn China faces material limits and the impact of US sanctions; catching up to the latest cutting edge may take a decade or more.
Beijing’s policy signals a shift in emphasis. The Communist Party’s recent Five‑Year Plan downplays explicit ambitions for chip dominance and instead highlights AI and a “model‑chip‑cloud‑application” framework that treats advanced chips as part of a broader computing ecosystem. China is prioritizing pragmatic, task‑oriented AI for industry that requires less compute power — workloads domestic chips can handle cost‑effectively.
That cost advantage is important. Chinese chips and AI systems often trade peak performance for much lower cost, boosting adoption across the Global South where governments and firms increasingly prefer Chinese solutions. Trendforce reported Chinese AI platforms — including DeepSeek and Alibaba’s Qwen — captured roughly 15% of the global AI model market by late 2025. That expansion poses a long‑term challenge to US tech giants such as Microsoft and Google, even as they plan record spending on AI infrastructure.
The US lead in AI and chip manufacturing faces other practical limits. ICIS warned that US data centers, which depend on high‑end chips, could be constrained by the nation’s strained power grid. By contrast, China’s expanding power sector offers spare capacity that could support large-scale data‑center deployment; ICIS projects an estimated 400 gigawatts of spare capacity by 2030. Cheap energy helps offset relative chip inefficiency for AI and other compute‑heavy applications.
ICIS outlines three broad outcomes: the US remedies its power‑grid weaknesses and maintains its lead; the US continues to dominate advanced chip research while Chinese AI systems proliferate in the Global South; or escalating trade and geopolitical tensions split the market into two separate AI ecosystems. Regardless of the path, established chipmakers face a future in which Chinese competitors underprice them while rapidly narrowing gaps in sophistication and reliability.
Edited by: Tim Rooks