Successful Conclusion of the 9th Digital China Summit Forum on AI Industry Development

On April 28, the important sub-forum of the 9th Digital China Construction Summit, themed “AI Industry Development and Empowering New Industrialization,” was held at the Fuzhou Strait International Conference and Exhibition Center. This forum was hosted by the Ministry of Industry and Information Technology and co-organized by the China Academy of Information and Communications Technology, Inspur Yunzhou Industrial Internet Co., Ltd., and Nanjing New Generation Artificial Intelligence Research Institute. It gathered top minds from government, industry, academia, and research to discuss the deep integration of AI and manufacturing.

The forum was themed “Digital Intelligence Leading New Industry, Modulus Resonance Strengthening the Real Economy,” closely following the deployment of the “Implementation Opinions on the Special Action of ‘AI + Manufacturing.’” It focused on key tasks such as data governance, the construction of high-quality data sets, the research and development of industrial large models, and their application in real-world scenarios, systematically showcasing China’s latest achievements and forward-looking layouts in empowering new industrialization with AI.

Strategic Path of Modulus Resonance

In his speech, Du Guangda, Deputy Director of the Science and Technology Department of the Ministry of Industry and Information Technology, emphasized that the core of AI empowering new industrialization lies in solving the two major issues of “technology implementation” and “value creation.” The joint issuance of the “Implementation Opinions on the Special Action of ‘AI + Manufacturing’” by the Ministry of Industry and Information Technology and seven other departments has established “modulus resonance” as a key action for the intelligent transformation of the manufacturing industry. He stressed that this is not only a choice of technical routes but also a reshaping of industrial logic. In the future, efforts will focus on solidifying a unified and authoritative benchmark testing system for industrial large models, overcoming common technologies for the collaborative evolution of data and models in key industries such as chemicals, equipment, and electronics, and building a number of public service platforms to promote the transition from pilot demonstrations to large-scale replication.

Wei Liang, Vice President of the China Academy of Information and Communications Technology, stated that the core essence of “modulus resonance” is to build a virtuous cycle of “high-quality data - high-performance models - high-value applications.” He revealed that a notice on jointly implementing the “Modulus Resonance” action for 2026 will be released soon, further clarifying the specific paths for the collaborative promotion of AI models and data resources. The China Academy of Information and Communications Technology will continue to play a bridging role under the leadership of the Ministry of Industry and Information Technology, ensuring a complete closed loop of “data - model - application” to solidify the core foundation for the digital transformation of the manufacturing industry.

Milestone Achievements Unveiled

The forum witnessed the release of a series of milestone achievements, with “modulus resonance” being the most prominent keyword. In the presence of leaders from the Ministry of Industry and Information Technology and the National Data Bureau, the “Modulus Resonance” action plan was officially launched, marking a new phase in the systematic promotion of the deep integration of AI and manufacturing in China.

Wei Liang then provided an in-depth interpretation of “modulus resonance” and released the “Artificial Intelligence Modulus Resonance Research Report (2026).” This report systematically outlines the theoretical framework, industrial paths, and practical cases of modulus resonance, providing authoritative guidance for the industry.

In terms of evaluation system construction, the revamped community “Fangsheng” large model benchmark testing capability was officially launched, initiated by Feng Junlan, Chief Scientist of China Mobile, and Wei Liang. This system aims to establish a unified, authoritative, and reproducible evaluation standard for industrial large models, addressing the challenges of heterogeneous scenarios, indicators, and data sets. The forum also released the results of the “Fangsheng” series of tests and a series of thematic reports on AI, showcasing China’s cutting-edge progress in AI evaluation, research, and application ecosystem construction.

Empowering Various Industries with Cutting-Edge Practices

The chemical industry is one of the pillars of the national economy. Ye Mao from the Dalian Institute of Chemical Physics of the Chinese Academy of Sciences pointed out that traditional chemical research and development has long been constrained by issues such as long cycles and high investments due to “gradual scaling up.” The team has built a full-chain big data center for petrochemical and chemical processes, releasing an “intelligent chemical large model” with deep mechanism reasoning capabilities, and has established a full-chain intelligent application group around enterprises, leading the construction of a national-level “full-industry collaborative ecosystem” with the support of the Ministry of Industry and Information Technology, aiding the chemical industry in transitioning towards low-carbon, high-end, and intelligent development.

Qi Guangpeng, Vice President of Inspur Group and Chairman of Inspur Yunzhou, presented on “AI + Industrial Internet Creating a New Paradigm for Empowering New Industrialization,” proposing that “AI + Industrial Internet” is the strategic infrastructure for new industrialization. He shared Inspur Yunzhou’s layout in industrial large models, industrial intelligent bodies, and industrial embodied intelligence, vividly illustrating the dual-driven model of “new technological supply leading” and “new industrial paradigm empowering” through cases such as high-end intelligent grinding machines from Jinan Heavy Industry and industrial large model craftsman factories for electromechanical pumps, transforming the digital divide into accessible pathways.

Xian Xiaoyu, Deputy Director of the Technology Research Department of CRRC Industrial Research Institute, elaborated on the construction of AI modulus resonance capabilities in the field of rail transit equipment, revealing how CRRC builds a “wheel-cutting” large model system and a high-quality data set as dual foundations, creating a full lifecycle AI closed loop covering research and development, manufacturing, and operation and maintenance.

Revolutionizing Paradigms: Reshaping Industrial Logic with Modulus Resonance

Li Sun, Deputy Director of the Platform and Engineering Department of the AI Research Institute of the China Academy of Information and Communications Technology, systematically explained the “New Paradigm of AI Industry Application - Modulus Resonance System.” He pointed out that the industry currently faces prominent issues such as the disconnection between model capabilities and business scenarios, insufficient release of data value, and fragmentation of technical systems. The modulus resonance system, centered on high-quality data sets, efficient models, and high-value applications, constructs a virtuous cycle mechanism connecting data-driven model evolution, model empowerment of application innovation, and application feedback on data accumulation. It is a key link connecting data governance, algorithm innovation, and industrial digital transformation, serving as an important engine for releasing the multiplier effect of AI and cultivating new productive forces.

Wang Lu, Chairman of Meilin Data Technology Co., Ltd., proposed the “One Body and Two Wings” AI strategy, emphasizing that the engineering implementation of AI requires the integration of information and digitalization achievements to form high-quality data sets, promoting deep integration of technology and business, and accelerating the replicability and cost controllability of AI applications.

Han Han, Chairman of Beijing Zhongshu Ruizhi Technology Co., Ltd., proposed a “Chinese solution” from the perspective of industrial intelligent bodies. She pointed out that in the future, the core of national industrial productivity competition lies in the combat effectiveness of intelligent body legions and data refining capabilities. Zhongshu Ruizhi utilizes the industry-leading “knowledge self-evolution” core technology to create a full-stack infrastructure of six intelligent engines, achieving a full-link autonomous closed loop of “perception - diagnosis - decision - execution” in key areas such as energy, military industry, and communications, supporting the construction of new industrialization with controllable core technologies.

Collaborative Exploration: From Pilot to Large-Scale Modulus Resonance

In a high-end dialogue session, experts from the China Academy of Information and Communications Technology, CRRC Industrial Research Institute, Dalian Data Industry Co., Ltd., Liying Smart (Beijing) Data Technology Co., Ltd., and Shenxinfu Technology Co., Ltd. engaged in in-depth discussions on core topics such as the construction of high-quality industrial data sets, the establishment of a unified large model evaluation system, the bidirectional resonance mechanism of modulus, and the path to large-scale replication.

The dialogue focused on four key propositions: How to build an industrial high-quality data set system to address issues of low data quality, difficult labeling, compliance challenges, and reuse difficulties? How to establish a unified, authoritative, and reproducible benchmark testing system for industrial large models? How to understand the bidirectional resonance mechanism of “using models to drive data and using data to empower models” to break through the bottlenecks of industrial scenario implementation? How can industrial modulus resonance transition from pilot demonstrations to large-scale replication, establishing quantifiable value assessments and replicable paths?

Experts unanimously agreed that modulus resonance is not only a technical issue but also a systemic engineering challenge. The current biggest bottleneck lies in the insufficient collaboration between data governance and model adaptation. In the future, efforts need to focus on standards, mechanisms, and ecosystems to accelerate the construction of a sustainable “data - model - application” closed loop, promoting modulus resonance from “bonsai” to “landscape.”

Building a Strong Manufacturing Nation with Modulus Resonance

The successful holding of this forum marks a solid and critical step for China on the road to empowering new industrialization with AI. “Modulus resonance,” rooted in data, winged by models, and centered on applications, sees all parties in government, industry, academia, and research working together to build a virtuous cycle ecosystem, promoting efficient allocation of data elements and deep release of model capabilities, leading new industrialization with digital intelligence and strengthening the real economy with modulus resonance.

Digital intelligence leads new industry, and modulus resonance strengthens the real economy—this gathering of wisdom and consensus has written a vivid chapter in accelerating the construction of a strong manufacturing nation, a strong network nation, and a digital China.