Weekend Long Read: China’s Chip Strategy for the AI Era Is a Bet on Disruptive Innovation

[para. 1] In recent years, the United States has intensified efforts to curb China’s development in semiconductors, driven largely by a broader strategic competition over artificial intelligence (AI) dominance. This geopolitical rivalry marks AI as not only a significant technological leap but also the culmination of a century-long human-machine convergence, focusing on who controls the next phase of cognitive technology[para. 2][para. 3]. Central to this race is the development and control of semiconductors, which form the foundation for AI advancement. The US aims to keep China at least two generations behind in chip manufacturing by coordinating with key allies to restrict exports of advanced equipment, leading to the fragmentation of global semiconductor supply chains[para. 4][para. 5].

[para. 6] Despite these pressures, China's semiconductor sector remains robust, being the only country besides the US with comprehensive capabilities across the full semiconductor value chain. China’s strategic approach has progressively evolved through four key stages: initial domestic development under US restrictions; opening up and reliance on imports post-1979 reforms; a pivot to self-reliance following geopolitical incidents in 1999; and an accelerated push for domestic innovation after US containment efforts intensified from 2018 onward[para. 9][para. 10][para. 11][para. 12][para. 13]. Notably, from 2018 to 2024, revenues in major Chinese chip sectors surged—chip design grew 156%, outsourced semiconductor assembly and test (OSAT) grew 43%, and chip manufacturing increased 144%[para. 14].

[para. 15][para. 16][para. 17] The US has consistently imposed multilateral and unilateral sanctions, including historical embargoes and more recent measures such as the Entity List, the CHIPS Act, and sweeping export controls. In turn, China has doubled efforts to establish a self-sufficient chip ecosystem, with major regional hubs emerging in areas like Beijing, Shanghai, and Guangdong’s Greater Bay Area. As of 2023, two Chinese companies ranked in the world’s top 10 fabless semiconductor firms, and several foundries and OSAT companies also placed highly on global lists[para. 20]. By 2024, China’s integrated circuit (IC) industry sales exceeded 1.43 trillion yuan ($201 billion), with remarkable growth in the fabless and manufacturing sectors[para. 21][para. 22][para. 23].

[para. 24][para. 25][para. 26] The US pressure campaign, particularly since 2018, has paradoxically triggered rapid growth and innovation in China’s semiconductor equipment and materials sectors, which have experienced annual growth rates above 20% since the late 2000s. Furthermore, half of the chips consumed in China are now domestically produced, up from 13.5% in 2013—an indicator of increasing self-reliance[para. 27].

[para. 28][para. 29][para. 30][para. 31][para. 32][para. 33] The race for dominance in AI technology has made it crucial for China to forge new, disruptive technological paths. While deep learning remains the cornerstone of AI research, gaps remain in matching human cognitive abilities. China’s large AI models currently depend heavily on US-made chips, making technological independence vital. The coming years may see China develop alternative chip and AI pathways, aiming for breakthroughs that mimic or surpass human cognition[para. 34][para. 35][para. 36].

[para. 37][para. 38][para. 39][para. 40][para. 41][para. 42][para. 43][para. 44] The article underscores that while the US holds clear leads in advanced process nodes, China’s strengths lie in resources for building large-scale, less advanced computing platforms. The limitations of Moore’s Law suggest that future advancements will come from architectural and system-level innovations. Cultural and linguistic diversity means that AI development is inherently local and national models have distinct advantages.

[para. 45][para. 46][para. 47][para. 48][para. 49] Talent development is another key front, with China's improved working environment narrowing the gap in expertise and compensation. While China’s engineering capabilities now rival those of the US, Europe, and Japan, a shortfall in original, foundational innovation persists. However, China’s market is adept at rapidly scaling new technologies, particularly in AI-driven applications such as new materials and pharmaceutical discovery. These fields benefit from AI’s capacity to shorten research cycles and facilitate breakthroughs—strengthening China’s prospects for leadership in emerging technology domains.

[para. 50] The article is authored by Wei Shaojun, vice chairman of the China Semiconductor Industry Association and professor at Tsinghua University, based on his remarks at the 3rd Asia Vision Forum.