China's AI development: A distributed and experiment-driven path

Editor's note: Lin G. is a CGTN economic commentator. The views expressed in this article are the author's own and do not necessarily reflect those of CGTN.

In the ongoing global race to shape the next wave of technological innovation, artificial intelligence (AI) has emerged as a central arena where countries seek to secure future advantages. Unlike previous major technological undertakings in China, such as aerospace or high-speed rail, which relied on concentrated scientific teams and top-down project management, AI is evolving under a markedly different dynamic.

In China, the development of AI has taken on a distributed, multi-point approach, characterized by widespread experimentation, localized support mechanisms, and strong integration with educational and research institutions. This approach reflects the unique characteristics of AI as a technology — its rapid innovation cycles — while simultaneously highlighting China's adaptive institutional environment.

A collaborative ecosystem of local governments, universities, and startups

At the core of this phenomenon is the interplay between local governments, universities, and entrepreneurial teams. Across the country, regional governments actively seek to foster AI-related industries within their jurisdictions. This is not the result of a centralized directive, but rather a natural outcome of inter-local competition.

Each locality, aiming to strengthen its technological and industrial base, provides various forms of support to AI projects. This support often includes funding initiatives, guidance in navigating administrative and legal requirements, and facilitation of partnerships between startups and universities.

The intention is not to dictate the direction of AI research or to preselect winners; instead, it is to create an enabling environment where research teams can focus on core technological challenges without being encumbered by peripheral operational hurdles.

Universities play a complementary role in this ecosystem. In many regions, leading universities collaborate closely with local entrepreneurial projects, contributing research talent, technological expertise, and incubation support. Talented researchers, motivated by the prospect of entrepreneurship and the chance to contribute to cutting-edge technology, often launch startups.

This organic interaction between academia, entrepreneurship, and local administration has created a dense network of innovation that spans across provinces, cities, and industrial sectors. It also ensures that promising ideas have multiple avenues to develop, test, and potentially scale, rather than being funnelled into a single top-down initiative.

Small teams, big variety: How China conducts AI experiments everywhere

One of the defining features of China's AI development is its contrast with the "big tech" model more prevalent in developed countries. While large technology companies in China, such as Alibaba, Tencent, Baidu, and ByteDance, maintain substantial AI research teams and develop significant products, the broader innovation landscape is not solely dominated by these major players.

A multitude of smaller startups, often ranging from a few dozen to over a hundred employees, are actively pursuing AI projects in parallel. These companies benefit from the local ecosystem described above, which reduces their operational friction and allows them to experiment freely with novel technological approaches.

Unlike the traditional narrative of startup absorption by major corporations, many of these small companies retain independence, supported by local governments and universities, allowing for a diversity of technological exploration that might otherwise be curtailed by corporate consolidation pressures.

This distributed model has produced several practical advantages. AI, unlike more capital-intensive or hardware-bound sectors, benefits from variety in experimentation. Different teams explore different technological pathways, architectures, and applications, increasing the likelihood that innovative breakthroughs emerge.

The interaction between large tech companies and smaller startups is also shaping innovation pathways. While the large companies continue to invest heavily in AI research and maintain significant market presence, they coexist with a network of smaller innovators, rather than absorbing them wholesale. This coexistence preserves a multi-directional innovation landscape, where entrepreneurial creativity is not stifled by corporate hierarchies or dominant platforms.

The current stage of AI development in China is primarily exploratory; large-scale commercial applications are still expanding, and the optimal technological paths are not yet determined. By maintaining a broad base of innovators, China can pursue multiple experimental avenues at the same time, increasing the chances of discovering solutions that are both technically robust and commercially viable.

Historical experience has provided a useful reference point. In the early stages of China's renewable energy sector, particularly in solar photovoltaic technology and battery development, a similar pattern of distributed experimentation occurred. Numerous regional initiatives, universities, and private companies explored different technological approaches simultaneously. While some efforts did not lead to scalable solutions, others ultimately produced globally competitive technologies.

Importantly, in many cases, the paths pursued in China differed from those in the United States, demonstrating the value of parallel experimentation and diversity of approaches. This experience has shaped expectations for AI development: Fostering multiple experimental nodes increases resilience and broadens the range of potential breakthroughs.

Strong societal awareness of AI with China's 15th Five-Year Plan recommendations

In addition to technological experimentation, China's AI ecosystem benefits from a strong societal and market awareness of AI. Under the guidance of China's 15th Five-Year Plan recommendations, which emphasize accelerating high-level technological self-reliance, the AI ecosystem has been developing within a broader strategic framework. The new Five-Year Plan recommendations underscore the integration of education, technology, and talent development, aiming to enhance autonomous innovation capabilities and promote new quality-driven productivity.

This widespread understanding has fostered a culture of experimentation, where AI is not merely a technical domain but a tangible tool with applications in finance, healthcare, manufacturing, logistics, and urban management. Awareness and enthusiasm for AI create a dynamic feedback loop: As research teams experiment, they generate real-world use cases that inform further research.

The role of external pressures, including technology restrictions from the US, also influence this ecosystem. China's AI development cannot rely on external technological sources, as access may be restricted or uncertain. Consequently, local governments, universities, and startups naturally emphasize homegrown technological solutions, fostering autonomous innovation.

This emphasis is reactive to external realities rather than the result of an overarching central mandate. It is a pragmatic response to a changing global technological landscape, reinforcing the importance of multiple experimental pathways and decentralized innovation networks.

Conclusion: Building an ecosystem for AI growth in China

As AI continues to evolve globally, China's distributed, trial-and-error model allows it to draw on a wide range of testing to advance its AI development. It is not the product of a single master plan, but the emergent result of competitive dynamics, development incentives, and widespread entrepreneurial energy. By supporting multiple pathways for innovation, integrating academia and local governance, and fostering a societal culture attuned to AI, China is generating the conditions under which new technological paradigms can emerge, evolve, and scale.