At the beginning of 2026, the development of industrial intelligent agents will enter a "golden period" of policy and industry resonance. The Ministry of Industry and Information Technology and eight other departments jointly issued the "Implementation Opinions on the Special Action of 'Artificial Intelligence+Manufacturing'", which clarifies the core goal of launching 1000 high-level industrial intelligent agents before 2027, injecting strong impetus into industrial development. In this context, leading manufacturing companies and industrial chain enterprises such as Midea and Haier are accelerating their layout and promoting the implementation of intelligent manufacturing with a full scenario intelligent agent matrix as the core. The industry believes that the sustained release of policy dividends and the deep implementation of enterprise practices form a two-way empowerment, promoting the transition of industrial intelligent agents from single point pilot to large-scale popularization, and becoming the core driving force for the transformation of manufacturing industry from "scale expansion" to "quality leap" and the cultivation of new quality productivity. By the beginning of 2026, the field of industrial intelligent agents will undergo a triple policy increase, including national top-level design, local implementation rules, and the setting of the two sessions. In January, the Ministry of Industry and Information Technology and eight other departments jointly issued the "Implementation Opinions on the Special Action of 'Artificial Intelligence+Manufacturing'", proposing to promote the deep application of 3-5 general large-scale models in the manufacturing industry by 2027, form distinctive and fully covered industry large-scale models, launch 1000 high-level industrial intelligent agents, create 100 high-quality datasets in industrial fields, and promote 500 typical application scenarios. At the same time, many cities such as Shenzhen and Chongqing quickly introduced supporting policies after the start of the new year, focusing on funding incentives, computing power support, scene openness, and other directions to provide precise guarantees for the landing of industrial intelligent agents. On March 5th, the government work report proposed to deepen the expansion of "artificial intelligence+" and promote the accelerated promotion of the new generation of intelligent terminals and agents. Du Yanze, Senior Research Manager of International Data Corporation (IDC) China, said in an interview with Economic Reference News that "industrial intelligent agents are becoming the core lever of 'AI+manufacturing', fundamentally because their role has shifted from 'passive tools' to' autonomous digital employees'. Currently, more and more companies are paying attention to the input-output ratio of AI, with the core demand shifting from 'whether to do AI' to 'how AI can truly help improve quality, reduce costs, and increase efficiency'. ”The 2025 survey of Chinese industrial enterprises released by IDC shows that the proportion of enterprises that have applied large models and intelligent agents has rapidly increased from 9.6% in 2024 to 47.5% in 2025; Among them, the proportion of enterprises that simultaneously apply in multiple links such as research and development, manufacturing, and supply chain has also jumped from 1.7% to 35%. This means that industrial intelligent agents are moving from single point experiments to cross link collaborative applications. Experts believe that policy empowerment will accelerate the penetration of intelligent agents throughout the entire value chain, including research and development, production, and supply chain. This will drive the manufacturing industry from "automation" to "autonomy", giving rise to new forms of production such as black light factories and flexible manufacturing, and reshaping the competitive advantage of the industry. IDC predicts that by 2028, the scale of AI expenditure by Chinese industrial enterprises will approach 90 billion yuan, with a compound annual growth rate of 37.7%. This means that AI has entered the "scalable expansion period" from the "concept investment period". From the perspective of growth range, with the continuous promotion of policies and the gradual clarification of enterprise investment return on investment (ROI), it is relatively reasonable and sustainable for China's "AI+manufacturing" market to maintain a compound annual growth rate of around 35% in the next 3 to 5 years. With the continuous advancement of the "AI+manufacturing" industry chain, from manufacturing leaders to industry chain enterprises, they are all accelerating the implementation of intelligent manufacturing with a full scenario intelligent agent matrix as the core. In January 2026, Midea's subsidiary Meiyun Zhishu released the Meiqing AIGC3.1 platform and intelligent agent factory solution, building an intelligent agent matrix covering the entire value chain of research and development, manufacturing, supply chain, marketing, and management, with a total of 158 core scenarios implemented. Midea claims that in the manufacturing sector, using intelligent agents such as TPM and molds to predict equipment failures and optimize processes can help improve OEE (overall equipment efficiency) by 30% and double inspection efficiency; On the supply chain side, intelligent agents have achieved a 39% reduction in end-to-end delivery cycles and a 30% reduction in inventory turnover days. Haier Smart Home also announced that it will achieve full AI automation through the super intelligent agent "Smart Small Energy", with dual track technology creation and application. The research and development efficiency will be improved by 90%, procurement costs will be reduced by 10%, and office efficiency will be increased by 80%, promoting a new work mode of collaborative evolution between humans and AI. Shenzhen technology company xTool, which focuses on the laser engraving machine track, officially released the world's first AI creative intelligent agent - AImake - in January. As the industry's first intelligent agent with "manufacturing context perception", AImake has achieved an instant transformation from natural language creativity to machinable design drawings, greatly reducing the technical threshold for non professional users to enter the field of laser manufacturing. In March, Huawei focused on solving the difficulties of AI intelligent agent landing and launched an AI data platform, laying a solid data foundation for enterprise digital transformation. In addition, listed company Weston recently revealed on an interactive platform that its industrial AI intelligent agent software can meet the needs of industrial enterprise application scenarios. It has changed the human-computer interaction mode of software products, further promoted product technology upgrades, and enriched the company's product line. Hande Information stated that the company has a self built "Deling" B-end AI application system, among which the "Lingshou" business intelligent agent series, including AI intelligent agents for various business scenarios such as manufacturing, marketing, finance, supply chain, personnel, and comprehensive operation enterprises, are gradually being implemented in the actual scenarios of top customers. The industry believes that with policy support, industrial intelligent agents are accelerating their implementation and promoting the transformation and upgrading of intelligent manufacturing. However, in the process of large-scale implementation, there are still multiple challenges such as technological adaptation, data infrastructure, cost-effectiveness, and ecological synergy, which are intertwined and become key obstacles to their transition from pilot to popularization. Industry experts point out that the development and deployment of industrial intelligent agents require multiple investments such as computing power, algorithms, and talent. There are pain points such as high technical barriers, large capital investment, and high data security risks, including the cost of training computing power for industrial large-scale models, customized development costs, and professional operation and maintenance personnel costs, which impose a heavy burden on small and medium-sized enterprises. In the short to medium term, the most core and urgent challenge facing the upgrading of the 'AI+manufacturing' industry still mainly comes from the 'market side' rather than the 'technology side'. ”Du Yanze stated that on the one hand, Chinese manufacturing companies' awareness and enthusiasm for AI continue to rise, and policy support, open-source technology ecosystem, and talent supply are also constantly improving; On the other hand, enterprises generally face the constraints of profit margins and tight budgets, which makes their willingness to pay for AI applications and investment pace more cautious. Specifically, Du Yanze believes that the current core constraints are mainly reflected in the following aspects: the pressure on enterprise ROI is high, and AI projects must reflect business value faster and more clearly; The market competition is too fierce, and homogenization schemes and price competition have compressed the sustainable investment ability of manufacturers; The supply-demand matching is still insufficient, and products and services that truly "understand intelligence and are familiar with the industry" are still scarce. In addition, some industry insiders have expressed that some manufacturing enterprises, especially small and medium-sized enterprises, have problems with fragmented industrial data, low standardization, and non-standard security management, which make it difficult to meet the needs of AI model training and application; At the same time, the standard system and evaluation system of industrial AI have not been fully unified, and the system interfaces and data formats of different enterprises are incompatible, which increases the integration cost and implementation difficulty of AI applications. (New Society)