Abstract

In the context of the digital economy, high-quality artificial intelligence data serves as a core production factor for the large-scale model industry. In contrast, data pricing is a key mechanism for revitalizing the domestic data factor market and balancing compliance constraints with the value realization of the AI industry. China has established a distinctive domestic data governance framework characterized by public data authorization, recognition of data assets on balance sheets, and tiered classification supervision. The operational logic, transaction scenarios, and regulatory rules of the domestic data market fundamentally differ from those in Europe and the Americas, giving rise to unique research topics distinct from those in other regions worldwide. To systematically trace the evolution of domestic research on data pricing, identify core local issues and distinctive key disagreements, and address the limitations of existing studies—particularly their narrow quantitative analysis dimensions and failure to account for practical constraints related to high-quality AI data supply—this paper selects 396 relevant publications indexed by CNKI from March 2015 to March 2025, focusing on the China data market as a sample. Utilizing the CiteSpace tool, we constructed temporal publication timelines, author collaborations, institutional partnerships, and multi-level knowledge graphs based on keywords to systematically analyze the overall landscape of domestic research and its academic collaboration ecosystem. The study identifies four prominent localized challenges in current domestic research: First, there is a lack of a unified valuation framework tailored to China's public-private dual data structure, with existing theoretical systems being fragmented; second, current pricing models primarily focus on general datasets, failing to address the full lifecycle requirements of AI data—including annotation, desensitization, rights confirmation, and compliance costs—with limited applicability across specific scenarios; third, theoretical frameworks remain disconnected from real-world transaction contexts such as domestic data exchanges and public data licensing operations, lacking empirical validation using local transaction data; fourth, collaboration among policy researchers, universities, and data trading platforms remains weak, leading to scattered research resources and frequent duplication of efforts. Addressing these gaps in localized research priorities and paradigms, this paper proposes targeted approaches across four dimensions—building localized theoretical frameworks, innovating AI scenario-specific pricing methods, implementing industry-specific solutions, and enhancing cross-stakeholder collaboration mechanisms—to provide theoretical foundations for developing data pricing mechanisms that align with China's institutional characteristics and serve the domestic AI industry.

Download

Share

COinS