数据要素市场化：制度、场景与AI的协同发力
This year's Government Work Report highlighted the 'accelerated release of data element potential'. This not only affirmed the work done in 2025 but also signaled a new push for 2026 to leverage data as a key driver of economic and social development.
From establishing a comprehensive policy system to promoting industrial data applications and enhancing AI development through high-quality datasets, a broad picture of fully releasing data element value is unfolding. 2026 marks the beginning of the '15th Five-Year Plan' and is positioned as the year for accelerating data element value release. The National Data Bureau, along with multiple departments, is working together to transform data into a powerful engine for high-quality development.
Data must first address the issues of 'supply, flow, utilization, and security'. This requires a robust set of foundational policies. According to Liu Lihong, Director of the National Data Bureau, over the past two years, the bureau has explored a '5+3+1' work framework, which includes building data infrastructure, fostering data integration, and advancing data industry.
The focus now is on implementing these policies, especially addressing concerns about data property rights. Liu emphasized that recent efforts have been made to clarify data ownership, usage, and operation rights through specific scenarios. The goal is to encourage all parties to share and use data confidently.
In the coming year, the construction of a unified national data property registration system and the release of policies for a nationwide integrated data market will be prioritized. These measures aim to integrate scattered data resources into a standardized market system, thereby solving the long-standing 'impossible triangle' of data security, compliance, and efficient circulation.
With the rise of AI technology, data plays an increasingly critical role. Wang Yanqing, Director of the Department of Information Technology Development at the Ministry of Industry and Information Technology (MIIT), pointed out that AI and other new-generation information technologies are rapidly integrating into every aspect of manufacturing, turning data into a vital production factor and strategic resource.
To achieve this, the MIIT has been promoting the 'two-in-one' integration, the Industrial Internet Innovation Development Program, and large-scale 5G application in industries and manufacturing sectors. By the end of 2025, the key process control rate of major industries had reached 68.6%, with many industrial-specific chips, modules, and terminal products being developed and applied.
This year, the 'Industrial Data Foundation Action' was launched, focusing on building high-quality industry-specific datasets for AI. Liu stated that by the end of 2025, more than 100,000 high-quality datasets had been established, with a total volume exceeding 890PB, equivalent to 310 times the digital resources of China's National Library.
Furthermore, the National Data Bureau is collaborating with 26 departments to select 72 leading units for high-quality dataset construction, 140 pilot units, and 104 exemplary cases. In the data annotation sector, seven cities—Chengdu, Shenyang, Hefei, Changsha, Haikou, Baoding, and Datong—are actively participating in the pilot construction tasks, with 47 outstanding annotation cases selected.
As a result of collective efforts, significant progress has been made in high-quality dataset construction. Liu revealed that by the end of 2025, the average daily token consumption had grown more than tenfold from early 2024 to March 2026, indicating the rapid growth of China's AI sector and its increasing competitiveness in global markets.
Looking ahead, the National Data Bureau plans to deepen the high-quality dataset construction action plan through six major initiatives: strengthening foundation, tackling annotation challenges, improving quality and efficiency, empowering applications, enhancing management services, and releasing value. These actions will be driven by scenario needs, further promoting pilot projects and creating AI-ready high-quality datasets.
With clear guidelines and practical steps, the National Data Bureau and the MIIT are converting the report's deployment into real-world practices. A thriving, efficient, and secure data market is taking shape, with its potential to drive innovation and high-quality development.