Digital News

　　The State-owned Assets Supervision and Administration Commission (SASAC) of the State Council recently released the first batch of 40 strategic and high-value artificial intelligence (AI) application scenarios of central state-owned enterprises (SOEs), advancing the construction of a “Strategic High-value AI Scenario Database for Central SOEs”.

　　“AI is accelerating the reshaping of the global industrial landscape. Promoting the deep integration of AI with the real economy is both a ‘compulsory course’ and a ‘necessary option’. Central SOEs must align with national strategies, closely follow frontier developments, leverage their advantages in resources and application scenarios, strike a balance between development and security, and make parallel efforts in both technology catch-up and application leadership. By mobilizing joint forces, we should collectively promote the high-quality development of AI.” Li Zhen, member of the CPC Committee and Vice Chairman of SASAC, emphasized the need to further advance the implementation of strategic high-value applications, strengthen top-level design, and deepen integration with the real economy. He highlighted the importance of identifying and building strategic high-value scenarios of great strategic significance, high economic returns, and close relevance to people’s livelihoods, while fostering application benchmarks in fields such as AI+science, biopharmaceutical, new material research and development, new industrialization, and embodied intelligence.

　　High-value application scenarios are the core carriers for AI to unleash its effectiveness. They define the limits of technological application value, provide fertile ground for the transformation of scientific achievements, and create key pathways for enabling industries with technology. The first batch of strategic high-value AI application scenarios released this time covers key areas including industrial manufacturing, smart energy, steel metallurgy, mineral exploration, materials research, pharmaceutical and healthcare, transportation and logistics, information and communications, smart construction, and smart agriculture.

　　China’s high-speed rail has achieved remarkable speed, with aerodynamic design serving as a critical core technology in train development. Addressing the need for large-scale aerodynamic simulations in high-speed train design, CRRC developed China’s first intelligent aerodynamic simulation foundation model for high-speed trains. The CRRC Zhuolun·Fengchi simulation model employs a data-driven modeling approach, trained and optimized using over 20 years of data accumulated through China’s independent development of high-speed electric multiple units (EMUs). This model replaces traditional flow-field simulations and simplifies the modeling process. Delivering accuracy comparable to conventional methods, it reduces simulation time from days on supercomputing resources to 10 seconds on a single GPU, improving overall efficiency by more than 30 times and advancing China’s aerodynamic design from experience-driven to data-driven paradigms.

　　Chen Jian, Deputy Director of the Digital Innovation Division, Department of Science and Technology Quality and Informationization of CRRC, noted that the Zhuolun·Fengchi simulation model features a fully independent and controllable technology stack. It has completed algorithm encapsulation, high-quality dataset construction, and verification of accuracy and efficiency. In the future, it will be applied in rapid aerodynamic performance evaluation and iterative design optimization for entire trains and components such as train noses, bogies, and cross-sections.

　　Helium is a vital strategic resource, and polyimide is a key raw material for high-purity helium separation membranes. Achieving independent development of this material is of great importance to alleviating China’s helium resource shortages. The Sinopec Beijing Research Institute of Chemical Industry has developed an intelligent polyimide material design platform through AI model training. This platform can generate new molecular design schemes, accurately predict gas separation performance and other parameters, and enable researchers to identify and validate optimal schemes. Its prediction accuracy for gas separation performance exceeds 90% on average.

　　China Railway Construction Corporation (CRCC), leveraging domestic computing power, hardware, and AI platforms from its proprietary AI computing center, has integrated AI with shield tunneling technology to build a new paradigm of intelligent construction characterized by multi-source data fusion, multi-process collaboration, and multi-model reasoning. Featuring eight functional modules, including intelligent equipment selection, intelligent construction organization, and intelligent shield diagnostics, this system has been successfully applied in over 20 projects, including the Ningbo-Zhoushan Railway, Chengdu-Chongqing Railway, and Shenzhen-Jiangmen Railway.

　　In addition, the “Intelligent Rapid Coal Quality Testing Technology Application Across the Entire Coal Industry Chain” of CHN Energy Investment Group employs multimodal sensing methods combining near-infrared, X-ray fluorescence, and laser technologies, achieving panoramic, authentic, accurate, rapid, and transparent coal quality testing. China National Biotec Group (CNBG) of China National Pharmaceutical Group Corporation (Sinopharm) has leveraged its “computing infrastructure + AI models + experimental validation” system on the Sinopharm Cloud to explore potential relationships among mRNA sequences. This system has supported sequence optimization for dozens of mRNA vaccines under its subsidiaries, accelerating the R&D of vaccines and innovative drugs, while advancing independent control of key technologies.

　　Regarding the promotion of strategic high-value AI applications by central SOEs, Liu Xingguo, Special Senior Research Fellow at the China Enterprise Confederation, analyzed that central SOEs can fully leverage their advantages of large-scale demand, comprehensive industrial support, and rich application scenarios. By prioritizing strategic significance, comprehensive value, and relevance to people’s livelihoods, they are driving participation of all parties in scenario exploration and application. Adhering to application-driven development, strengthening open collaboration and cooperation, and accelerating the application-guided growth to drive AI to empower all industries is conducive to promoting high-quality development.

　　According to reporters, the next step for SASAC of the State Council is to further advance the “AI+” special initiative of central SOEs. Based on the “Strategic High-value AI Scenario Database for Central SOEs”, upholding self-reliance and self-strengthening, emphasizing application-oriented development, and promoting the healthy and orderly development of AI, SASAC will continue to encourage SOEs to explore and open up core industry scenarios, foster collaborative building and sharing, and deeply integrate AI technological innovation with industrial innovation. By creating benchmarks in application, upgrading traditional industries, and opening new tracks in strategic emerging industries and future industries, SASAC aims to promote the transition from “bonsai-style” pilots to “rainforest-scale” applications.