At the 2026 Inclusion·Outer Temple Conference forum on "AI Economic Growth New Paradigms and Social Value Restructuring," Jiang Xiaojuan, a professor at the University of the Chinese Academy of Social Sciences and former Deputy Secretary-General of the State Council, stated that with China's enhanced innovation capabilities and the rapid development of AI technology, Chinese enterprises are transitioning from the traditional "domestic first, overseas second" globalization path to a "born global" native internationalization stage.

"In the AI era, all innovative products and services are inherently designed for a global audience from inception," Jiang said. She believes this shift is not merely about companies expanding overseas earlier, but rather that technology research and development, product design, business models, and even compliance systems must address the global market from the very beginning.

In the past, globalization for Chinese companies was often a gradual process. Companies would first build capabilities through technology introduction, absorption, and independent innovation, then complete product validation and scale within the domestic market, and only after achieving sufficient competitiveness would they enter overseas markets. Jiang summarized this model as "sequential globalization."

Jiang argues that as more Chinese industries reach the global frontier, companies now face "uncharted territory"—with no mature technologies available to continue importing or imitating, they must innovate independently, and once at the technological frontier, innovation inherently competes on a global scale.

"Artificial intelligence enables R&D foundation models and data to be reused and replicated at extremely low costs, and the network makes 'far away' and 'right next door' indistinguishable. Therefore, the marginal cost of global application is very low," she explained. Unlike the manufacturing era, where products were first strengthened domestically before entering overseas markets, AI products such as large models are typically global-facing at the moment of release.

Using open-source models as an example, Jiang cited data showing that this year, Chinese open-source models have surpassed the United States for the first time in download volumes on major global open-source communities; in July and August, the top six in the call volume rankings were all Chinese-developed open-source large models. She believes open-source models are lowering the barriers for different countries to access cutting-edge AI technology and significantly reducing the costs of cross-border flow and collaborative innovation of global innovation resources.