Representative End User Clients
Representative Automation Clients
Representative Software Clients
On April 28, 2026, the General Office of the Ministry of Industry and Information Technology and the General Department of the National Data Administration jointly issued the Notice on the Implementation of the 2026 “Model-Data Resonance” Initiative ([2026] No. 193), officially launching the 2026 “Model-Data Resonance” initiative. Based on the Opinions on Further Implementing the “AI+” Action and the Special Implementation Opinions on the “AI+ Manufacturing” Initiative, the policy coordinates industrial AI applications with data development, clarifying the implementation framework, seven core tasks, departmental responsibilities, and operational constraints. It is intended to address key industrial challenges, including the disconnect between data and AI models and barriers to the large-scale deployment of industrial AI.
I. Overall Implementation Framework and Core Tasks
The initiative covers 20 key national industrial sectors, including iron and steel, petrochemicals and chemicals, industrial machine tools, automotive manufacturing, power equipment, aerospace, semiconductors, and pharmaceuticals. Provincial-level authorities are required to promote implementation in no fewer than three industries, while central state-owned enterprises are required to select at least one industry for pilot application. The policy defines a clear three-phase implementation schedule: implementation plans are to be submitted by May 30, 2026; a midterm evaluation is to be conducted by August 30, 2026; and a final assessment is to be completed by the end of 2026. Standardized datasets, industrial models, and demonstration cases will be summarized and promoted nationwide, with the overall goal of building a closed-loop “data-model-scenario application” system and improving the broader industrial AI ecosystem.
The document specifies seven mandatory core tasks:
First, build high-quality, industry-wide datasets, develop industry-specific foundation models aligned with industrial processes and mechanisms, and promote standardized application cases.
Second, identify replicable, high-value industrial scenarios, construct scenario-specific datasets, and develop specialized AI agents for specific industrial applications.
Third, establish unified evaluation datasets and a standardized industrial AI evaluation system covering accuracy, reliability, compliance, and generalization.
Fourth, build a public collaborative “Model-Data Resonance” platform to support end-to-end data collection, labeling, model training, simulation verification, and deployment, reducing barriers to intelligent transformation for small and medium-sized manufacturing enterprises.
Fifth, establish cross-industry innovation consortia to coordinate joint research among manufacturers, computing service providers, data institutions, and research institutes.
Sixth, build regional digital transformation benchmarks in cities with strong smart manufacturing capabilities and data infrastructure.
Seventh, cultivate interdisciplinary talent with expertise in industrial processes, data analysis, and AI technologies; accelerate the formulation of national standards for industrial data and industrial models; and promote mature solutions and practices on a regular basis.
A collaborative governance mechanism involving the two departments is adopted. The Ministry of Industry and Information Technology takes the lead in coordinating industrial implementation, scenario and model R&D, and pilot city development. The National Data Administration is responsible for formulating industrial data standards, trusted data-sharing rules, and data security and compliance systems. The two departments jointly conduct supervision, evaluation, and final acceptance.
Three operational requirements are also defined: all data and AI applications must comply with data security and industrial confidentiality regulations; a “human-in-the-loop” mechanism requires AI to remain an aid to decision-making, with on-site operators retaining final operational authority; and existing PLC, MES, and other automated systems must remain compatible with the new architecture to enable phased, lower-cost digital transformation.
II. Overall Impact on Industry
As China’s first national policy designed to coordinate the development of industrial data and industrial AI, the 2026 “Model-Data Resonance” Initiative could significantly influence the pace and competitive dynamics of industrial AI adoption.
In the short term, it is expected to push manufacturers to address gaps in industrial data governance, positioning standardized, high-quality industrial data as a core production resource and helping move industrial AI from isolated pilots toward broader deployment.
In the long term, the industry could move away from fragmented approaches that prioritize models over data and toward a closed-loop model of “data governance-model optimization-scenario implementation.” The deeper integration of OT and IT systems, together with industry-specific vertical models and lightweight industrial AI agents, is likely to become an increasingly important technical direction, shifting factory operation and maintenance, quality inspection, and process optimization from reactive, post-event responses toward more predictive management.
III. Impact on Enterprises Across the Automation Value Chain
1. Industrial Control Hardware, Sensing, and Robotics Enterprises
First, market demand could expand. The policy requires comprehensive improvements in industrial data collection. As key data-acquisition devices and execution layers for AI-enabled applications, PLCs, servo systems, industrial sensors, gateways, and industrial robots could benefit from increased demand associated with the digital modernization of existing production lines and the construction of new smart factories. Industrial machine tool and CNC system enterprises could also benefit from demand for machine-tool datasets and process-specific AI agents.
Second, product requirements are likely to increase. Industrial equipment will increasingly need to support edge data preprocessing and edge AI inference, encouraging suppliers to upgrade intelligent controllers and machine-vision and sensing systems. Traditional hardware manufacturers relying solely on assembly and production could face increasing competitive pressure.
2. Industrial Software, Industrial Internet, and Industrial AI Platform Enterprises
These enterprises are likely to be among the primary beneficiaries of the initiative. MES, industrial data platforms, and industrial Internet platform vendors could serve as key participants in building public “Model-Data Resonance” platforms and industry datasets. They may also participate in developing industry foundation models and scenario-specific AI agents and in delivering digital transformation projects for government entities and central state-owned enterprises.
In addition, business models could continue to evolve. Revenue structures may shift from one-time software sales toward long-term value-added services, including platform subscriptions, model services, and AI agent operation and maintenance, creating opportunities for recurring revenue.
Industry differentiation could also intensify. Leading enterprises with deep industrial process expertise and full-stack data governance capabilities may take larger roles in innovation consortia, while smaller software providers with single-function products may need to collaborate with other companies to participate in major industrial projects.
3. Industrial Data Services, Data Labeling, and Data Security Enterprises
Short-term business opportunities could expand rapidly. Because building standardized industrial datasets ranks first among the initiative’s core tasks, demand could increase for the cleaning, labeling, and structuring of industrial time-series data, image data, and process-related text data, benefiting specialized industrial data governance and labeling providers.
Meanwhile, demand for compliant data services is also likely to increase. As the National Data Administration develops industrial data circulation and compliance systems, demand for privacy-preserving computing, industrial data encryption, and data-rights management services could grow.
4. Intelligent Manufacturing System Integrators
Regional benchmark projects could create new opportunities for system integrators. Integrators may undertake factory-wide model-and-data integration projects. Vertical industry integrators with extensive on-site scenario expertise may be better positioned to build proprietary datasets, develop scenario-specific AI agents, and participate in national demonstration projects.
Industry entry barriers are also likely to rise. Simple hardware integration may no longer be sufficient to meet policy requirements. Integrators will increasingly need data governance and model deployment capabilities, placing pressure on smaller firms that lack expertise in data and AI technologies.
5. Computing Infrastructure and Industrial Technology Providers
The construction of the “Model-Data Resonance” public platform will require industrial computing infrastructure, private industrial 5G networks, and industrial operating systems to support model training and simulation verification. Accordingly, computing service providers, industrial communication companies, and industrial OS vendors could benefit from supporting infrastructure projects, as well as related services such as industrial AI training and model evaluation.
Related ARC Insights
For additional ARC perspectives on industrial AI adoption, data foundations, and scalable deployment, see:
Beyond the Hype: How Industrial AI Pacesetters Are Rewriting the Rules of Scale in 2026 — examines how industrial organizations are moving beyond isolated AI pilots by strengthening data foundations and scaling operational AI.
AI in 2026: From Generative Tools to Agentic Digital Labor — explores the shift from generative AI toward agentic AI capable of executing more complex industrial and enterprise workflows.
Industrial AI's Role in Digital Transformation of Manufacturing Industries — provides ARC's broader framework for applying industrial AI across people, processes, technology, data, and governance.
Together, these ARC insights provide broader context for the “Model-Data Resonance” Initiative by highlighting the importance of trusted industrial data, scalable AI architectures, governance, and the operational capabilities needed to move industrial AI from experimentation toward broader deployment.