Identifying the Main Paths and Formation Mechanisms of Artificial Intelligence Technology Development in China

Dados Bibliográficos

ID22006859
AutoresJialu Ren (0009-0004-3906-3225, Xi'an University of Architecture and Technology), Yi Zhou (0000-0001-6130-687X, Xi'an University of Architecture and Technology, autor correspondente), Yong Zhou (0000-0003-0363-372X, Xi'an University of Architecture and Technology)
Ano2026
Volume16
Fascículo2
Data de publicação2026-04-01
Peer ReviewedSim
Open AccessSim
TipoARTICLE
PeriódicoSAGE Open (JOURNAL)
Identificadores do periódicoISSN: 2158-2440 • E-ISSN: 2158-2440
EditoraSAGE Publications (PUBLISHER • US)
DOI10.1177/21582440261432664
OpenAlexW7164809655
IdiomaEN
Referências citadas82

The rapid advancement and widespread adoption of artificial intelligence (AI) are profoundly reshaping numerous industrial sectors. Despite this, the primary evolutionary trajectories of AI technologies and the mechanisms governing their formation have yet to be comprehensively investigated. To address this knowledge gap, our research develops a patent citation network based on Chinese patent data spanning from 1985 to 2023. Initially, distinct technological communities are identified using community detection methods. Subsequently, the Search Path Count (SPC) algorithm and the critical path method are applied to uncover the main development trajectories within each community, enabling the tracing of AI’s evolutionary history in China and the forecasting of its future course. The study further employs Exponential Random Graph Models (ERGM) to scrutinize the formation mechanisms driving these paths, which are assessed against four dimensions of technological characteristics. The findings pinpoint eight principal development paths in AI. Each path exhibits unique characteristics concerning its technical domains, foundational technologies, as well as the key contributing cities and institutions, thereby reflecting the distinct roles and orientations of different technological communities. The ERGM analysis reveals that technological proximity, collaborative potential, and opportunities for recombination significantly and positively influence path formation, whereas knowledge diversity demonstrates a negative impact. Moreover, endogenous variables related to network structure are also identified as positive drivers for the emergence of these main development paths. Collectively, these findings illuminate the current trends and future directions of AI development. They offer crucial insights for technology developers and policymakers and provide actionable guidance for traditional industries navigating the AI-driven transformation

China · Complex network · Path dependence · Technological change · Technological evolution · Technology forecasting · Big Data and Digital Economy · Economic and Technological Innovation · Intellectual Property and Patents