On August 31st, the General Office of the Ministry of Industry and Information Technology (MIIT) issued a notice outlining a special initiative to nurture artificial intelligence application service providers. These providers, defined as enterprises or institutions offering services such as consultation, planning, deployment, operations management, and security governance for AI solutions, are the focus of this new campaign.
The initiative is part of broader efforts to implement the State Council's guidelines on deepening the "AI+" action plan, which specifically calls for cultivating AI application service providers and building a comprehensive AI application service chain. The MIIT is now organizing a special action to achieve these goals, with the following key objectives and tasks.
Where to begin
The primary objective is to establish a national resource pool of AI application service providers, enhancing their capabilities in technical innovation, integrated delivery, and security compliance. The goal is to cultivate a workforce of providers who deeply understand industry pain points, possess technical expertise, are aware of security risks, and excel in delivery and operations. By the end of 2026, the national resource pool is expected to include over 2,000 providers, forming a well-structured, multi-level echelon with clear division of labor and collaborative innovation. By the end of 2027, the pool should expand to at least 3,000 providers, supporting a fully collaborative, integrated, and comprehensive AI application service ecosystem.
Core initiatives
The first task involves building the provider resource pool. Regional authorities will survey and register local providers to create a database. By the end of 2027, provinces hosting national AI industry innovation pilot zones must have at least 100 providers in their local pools. The MIIT will aggregate these regional pools to form the national resource pool, which will be published periodically.
The second task focuses on improving service supply. Providers will be encouraged to lead "AI application service groups" in areas like model-data resonance and computing-power coordination, working with upstream and downstream partners to create integrated solutions. Regional innovation platforms, such as national AI pilot bases and tech business incubators, will be leveraged to support technology R&D, testing, and validation. Providers will also be supported in tackling key technical challenges like software-hardware adaptation and multi-model coordination, while enhancing their ability to integrate secure operating systems, databases, and AI training and inference chips. Security protocols for cybersecurity, data safety, and ethical governance must be embedded throughout the entire process, shifting from reactive to proactive risk management.
The third task aims to promote large-scale application adoption. Providers will be guided to develop modular, standardized solution packages and a range of "small, fast, lightweight, and precise" AI products and services for high-frequency, essential business needs. Through industry alliances, associations, and open-source communities, targeted supply-demand matching events will be organized to promote these providers. Real-world scenarios will be opened to providers for pilot implementations, fostering innovation and creating benchmark solutions. To accelerate adoption, mechanisms like first-purchase incentives and risk compensation will be explored to increase procurement of large models, intelligent agents, and Token services. Additionally, regions with the capacity will build comprehensive international expansion support systems, leveraging platforms like the Belt and Road Initiative and BRICS to facilitate the global deployment of high-quality AI projects.
The fourth task concerns strengthening support and safeguards. Providers will be encouraged to connect with national computing network hubs and use policy tools like "computing power vouchers" to reduce costs. Intelligent programming tools will be optimized to improve software delivery capabilities. Resources such as scenarios, data, models, and agents from various initiatives will be made available to providers through open interfaces or trial channels. Universities and vocational schools will be encouraged to partner with leading providers to establish training bases, cultivating interdisciplinary talent proficient in business, models, security, and delivery. Providers will also be urged to establish field deployment engineer (FDE) teams to support on-site implementation.
Implementation plan
Regional authorities must compile provider dossiers and submit their regional resource pools to the MIIT's Department of Science and Technology by December 1, 2026, with periodic updates to follow. Each region is required to establish at least ten "AI application service groups," each led by one provider and involving at least two upstream or downstream partners, with details submitted alongside the resource pool information.
The MIIT will establish a routine tracking and evaluation mechanism, organizing expert site visits to monitor progress and address challenges. Regional authorities must submit summary reports by December 1, 2027, detailing task completion. The MIIT will then assess the overall implementation and showcase successful providers, solutions, and service groups.
Organizational support
The MIIT will provide overarching guidance, promptly resolving critical issues during the initiative's implementation. Regions demonstrating effective results may receive preferential support in relevant policies and projects, with enhanced promotion of their achievements. An online platform for the national provider resource pool will be developed to facilitate information sharing. Local industry and information technology authorities are encouraged to implement supportive policies and allocate resources to assist these providers. Inquiries can be directed to the MIIT's Department of Science and Technology at 010-68205250.