The evolutionary mechanism of artificial intelligence industry collaboration networks: evidence from China

Collaboration networks play a crucial role in fostering innovation within the artificial intelligence (AI) industry. This study investigates the evolutionary mechanisms underlying collaboration networks within China’s AI industry. Utilizing a stochastic actor-oriented model (SAOM) framework, we integrate endogenous structural effects, exogenous organizational attributes, and dyadic proximity characteristics to analyze longitudinal patent data from 2013 to 2024. The results reveal that key network dynamics, such as transitivity and preferential attachment, actively shape tie formation. Importantly, universities and research institutions play a more central role in driving network evolution than firms, and organizations with higher innovativeness attract more collaborative partners. While geographical, cultural, and institutional proximities facilitate collaboration, technological proximity has a noteworthy negative effect, underscoring the importance of complementary knowledge in AI innovation. Our findings enhance the understanding of the differences in the different dimensional factors that drive the evolution of AI industry collaboration networks and advance the research on network dynamics.