Recently, the Guizhou Provincial Big Data Development Administration issued a notice to distribute the "Action Plan for Advancing the Construction of High-Quality Industry Datasets in Guizhou Province (2026-2028)."
The Plan requires taking the empowerment of artificial intelligence applications as the main line, with public data taking the lead in "stepping onto the field" as the driving force, and the construction of AI data factories as the key lever, leveraging Guizhou Province's comparative advantages, adhering to the principles of "application-driven and scenario-based, coordinated and rationally distributed, focused on key areas and innovative breakthroughs, compliant advancement and security assurance," implementing seven major initiatives: high-quality data supply, distinctive dataset construction, professional annotation breakthroughs, quality and efficiency improvement, application scenario empowerment, management service optimization, and accelerated value release, establishing and improving full-chain service capabilities for high-quality datasets spanning data supply, processing, circulation, and application, building a data element and artificial intelligence collaborative development ecosystem with Guizhou characteristics, and providing strong support for the construction of Digital Smart Guizhou.
The Plan specifies — By 2026, build more than 30 provincial-level key industry high-quality datasets, empower more than 20 AI demonstration scenarios, gather more than 30 data annotation enterprises, and have more than 20,000 people engaged in data annotation work.
By 2027, cumulatively build more than 60 provincial-level key industry high-quality datasets, empower more than 50 AI demonstration scenarios, build one AI data factory, gather more than 50 data annotation enterprises, and have more than 30,000 people engaged in data annotation work.
By 2028, cumulatively build more than 100 provincial-level key industry high-quality datasets, empower more than 100 AI demonstration scenarios, establish an AI data factory operational system, gather more than 300 dataset construction service ecosystem enterprises across the province, more than 100 data annotation-related enterprises, and have more than 40,000 people engaged in data annotation work.
The Plan proposes implementing seven major initiatives — high-quality data supply, distinctive dataset construction, professional annotation breakthroughs, quality and efficiency improvement, application scenario empowerment, management service optimization, and accelerated value release — comprising a total of 41 key tasks.
Implement the high-quality data supply initiative, focusing on the construction needs of high-quality industry datasets, vigorously advancing public data "stepping onto the field," and driving the development and utilization of enterprise data and personal data through classified and tiered approaches.
Implement the distinctive dataset construction initiative, focusing on 25 Guizhou-characteristic industry sectors including new comprehensive energy, new materials, deep processing of mineral resources, sauce-flavor baijiu, and advanced equipment manufacturing, establishing a "industry regulatory department + chain-leading enterprise + upstream and downstream enterprises + research institutes + data group + data service providers" collaborative working mechanism, following the "one chain, one dataset, one standard" approach, concentrating efforts in each segmented field or industry chain to build one high-quality industry dataset and develop one set of dataset construction standards, forming a batch of intensive, standardized, and professionalized key industry high-quality datasets.
Implement the professional annotation breakthrough initiative, promoting the clustering of the data annotation industry, building AI data factories, consolidating professional talent guarantees, encouraging collaborative innovation, cultivating and organizing a batch of professional practical annotation talent and expert teams, laying out and building data annotation parks through classified and tiered approaches, and achieving large-scale, standardized, and professionalized supply of high-quality datasets.
Implement the quality and efficiency improvement initiative, promoting the construction of functional platforms for high-quality dataset construction, strengthening common technology research and development, improving technical standards, expanding third-party professional evaluation, and effectively enhancing dataset quality levels.
Implement the application scenario empowerment initiative, building a "data flywheel" application system, adhering to application scenarios as the driving force, creating typical applications and pilot demonstrations of high-quality datasets, strengthening application innovation and supply-demand matching, and achieving model-guided data, data-empowered models, and model-data resonance.
Implement the management service optimization initiative, using data property rights registration as the lever, strengthening high-quality dataset management services, and building a distinctive dataset construction service system for Guizhou Province.
Implement the accelerated value release initiative, adhering to government guidance and market entity leadership, cultivating market consensus on paying for data, promoting the commercialization and assetization of datasets, and facilitating compliant, efficient, and secure circulation of datasets.
The Plan also proposes strengthening safeguard measures including organizational implementation, policy support, and security assurance. It is necessary to strengthen service efforts for key enterprise projects and expand dataset application scenarios; encourage innovative "exemption from application for immediate enjoyment" and "application for immediate enjoyment" fulfillment mechanisms to ensure policy benefits are delivered precisely and directly to market entities; strengthen classified and tiered data protection, improve data security risk assessment, monitoring and early warning, emergency response and other mechanisms, and guide business entities to strengthen industry self-discipline.
Comments