Summary of Results Released at the 2026 Artificial Intelligence Industry Conference: AIIA Releases 20 Results in a Concentrated Manner

From October 14 to 16, 2026, the 2026 Artificial Intelligence Industry Conference was held in Jinan. Yu Xiaohui, Member of the Party Leadership Group and Vice Minister of the Ministry of Industry and Information Technology; Zhou Naixiang, Deputy Secretary of the Shandong Provincial Party Committee and Governor of Shandong Province; Sun Ximin, Vice Governor of Shandong Province; Yu Haitian, Deputy Secretary of the Jinan Municipal Party Committee and Mayor of Jinan; Zheng Zhiming, Academician of the Chinese Academy of Sciences; and more than 800 guests from relevant departments, industry enterprises, and other units attended the opening ceremony. During the conference, a number of artificial intelligence research results were released in a concentrated manner, covering directions such as industry empowerment scenarios, technology evaluation, and standards systems, systematically presenting the latest judgments on the evolution of the artificial intelligence industry. Selected Research Results

1. "Map of High-Value Scenarios for AI Empowering New Industrialization"

Result introduction: Under the overall coordination of the Department of Science and Technology of the Ministry of Industry and Information Technology, the high-value scenarios for AI empowering new industrialization, jointly compiled by multiple research institutions and leading industry enterprises, were officially released. The high-value scenario map covers 20 sub-sectors and more than 1,000 high-value scenarios, comprehensively covering core areas of manufacturing. The map adopts the form of "one map for each industry, one file for each scenario" and builds a three-level system of "key breakthrough level, verification level, and promotion level," forming a reusable "living map" of industrial AI practice and providing path support for new industrialization. 2. "Artificial Intelligence Security Evaluation Framework for the Industry and Information Technology Field"

Result introduction: Under the guidance of the Cybersecurity Administration Bureau of the Ministry of Industry and Information Technology, led by the Artificial Intelligence Industry Alliance (AIIA), and jointly researched and compiled by 31 units including industry organizations, research institutions, and leading enterprises, the "Artificial Intelligence Security Evaluation Framework for the Industry and Information Technology Field (2026 Edition)" was officially released. Based on the needs of AI security evaluation in the industry and information technology field, the evaluation framework focuses on three dimensions: general security, application security, and business security; clarifies evaluation content, methods, and implementation requirements; provides reference for relevant units to identify security risks, assess security levels, and test the effectiveness of security protection; and promotes the standardized and systematic conduct of AI security evaluation work. Download link: Artificial Intelligence Security Evaluation Framework for the Industry and Information Technology Field (2026 Edition).pdf 3. "CSI BOC Shanghai-Shenzhen-Hong Kong Artificial Intelligence Industry Index"

Result introduction: With the support of relevant departments and bureaus of the Ministry of Industry and Information Technology and the People's Bank of China, the China Academy of Information and Communications Technology, together with Bank of China, the China Association for Public Companies, the Shanghai Stock Exchange, the Shenzhen Stock Exchange, and China Securities Index Co., Ltd., jointly released the "CSI BOC Shanghai-Shenzhen-Hong Kong Artificial Intelligence Industry Index." The index targets all listed companies in Shanghai, Shenzhen, and Hong Kong, strictly screening out 314 target enterprises that meet the standards, comprehensively covering key links in the industrial chain such as the foundation layer, framework model layer, and application layer. By introducing industrial chain tier weights and sector constraint mechanisms, it builds a listed company sample system that is both representative and trackable. The release of the index aims to further connect AI industry standards with the capital market and financial services, create a more standardized, transparent, and scientific industrial identification and evaluation tool, improve the precision and effectiveness of financial resource allocation, and promote better financial services for the high-quality development of the artificial intelligence industry. Download link: CSI BOC Shanghai-Shenzhen-Hong Kong Artificial Intelligence Industry Index.pdf 4. "Top Ten Keywords of the 2026 Artificial Intelligence Industry"

Result introduction: The top ten keywords of artificial intelligence have been released for five consecutive years since 2022. In 2026, from an industry perspective, the focus is on changes that can represent the annual mainlines in 2026, and along these mainlines, trends in the foreseeable short-term future are assessed. Through comprehensive expert surveys, statistics on hot events, and research by the writing team, ten keywords were selected. The top ten keywords unfold from three levels—technology, application, and ecosystem—from the inside out, jointly outlining a complete picture of the accelerated diffusion of artificial intelligence in 2026. Download link: Top Ten Keywords of the 2026 Artificial Intelligence Industry.pdf 5. "Embodied Intelligence Development Report (2026)"

Result introduction: Following the "Embodied Intelligence Development Report (2024)" and the "Embodied Intelligence Development Report (2025)," this is the third release of an embodied intelligence research report. This report focuses on new developments, new changes, and new challenges in the embodied intelligence industry since 2026, summarizes and sorts out the development characteristics of the embodied intelligence industry, technology, and applications at home and abroad, and provides an outlook on future development. Download link: Embodied Intelligence Development Report (2026).pdf 6. "Artificial Intelligence Standards Toolbox of the Artificial Intelligence Standardization Technical Committee of the Ministry of Industry and Information Technology"

Achievement introduction: First, intelligent mapping of the AI standards database, solving the problem of "searching." National, industry, and international standards are gathered in one database and accessible with one click; new terms are automatically extracted and updated on a regular basis; the terminology database and the standards database are mapped to each other, so similar standards and related terms can be grasped with one click. Second, the standard review Skill, solving the problem of "reviewing." We translated the review requirements of GB/T 1.1—2020 clause by clause into executable rules, developed the Skill, and introduced agents to assist with review, achieving evidence-based review, clear annotations, and improved quality and efficiency—each comment is labeled with the source of the rule and the textual evidence, and the conclusions are directly rendered as native document comments anchored to the specific text to be modified. Review accuracy reaches 80%, and the time required per document has been shortened from half an hour to 10 minutes. Its deeper value lies in transforming the experience in the minds of a few experts into reusable, accumulable industry common knowledge. Third, the standard drafting assistance plugin, solving the problem of "writing." Relying on the aggregated support of GB/T 1.1—2020, the standards database, and the terminology database, it provides full-process assistance with modular drafting, controllable progress, and clear formatting, with traceable revision marks and comparable version differences. The effect we hope to achieve is to lower the threshold for standard development so that everyone can become a standards expert. 7. "Guidelines for the Classification of the Artificial Intelligence Industry Chain"

Achievement introduction: In order to fully build consensus on the development of the artificial intelligence industry and better help various industry entities make precise policies, strategic layouts, and scientific investments, the China Academy of Information and Communications Technology (CAICT), together with Shanghai Guotou Xiandao Private Equity Fund Management Co., Ltd., Zero2IPO Research Center, Industrial Securities, China Mobile (Shanghai) Industry Research Institute, and more than 10 other units, jointly compiled the "Guidelines for the Classification of the Artificial Intelligence Industry Chain" specification, systematically sorting out and classifying the industry chain, and building a comprehensive, scientifically classified, and dynamically updated industry chain classification system. It divides the artificial intelligence industry chain into five layers: the energy layer, data layer, computing power layer, model layer, and application layer, and systematically sorts out the key technology nodes and key indicators in the key links of each layer of the industry chain, guiding local governments and institutions to accurately assess trends in industrial capital flows, better promoting two-way empowerment between industry and capital, and advancing the high-quality development of the artificial intelligence industry. 8. "Guidelines for Building High-Quality Datasets for Agent Capability Construction (2026)"

Achievement introduction: The "Guidelines for Building High-Quality Datasets for Agent Capability Construction (2026)" was launched in April 2026 and jointly compiled by the Artificial Intelligence Institute of the China Academy of Information and Communications Technology and Lenovo (Beijing) Co., Ltd., with the active participation and support of more than 30 units including Neusoft. The guidelines focus on the new requirements that agent capability construction places on high-quality datasets, systematically summarize dataset construction ideas and practical experience, and provide useful reference for the development of the agent industry and the construction of high-quality datasets. Download link: Guidelines for Building High-Quality Datasets for Agent Capability Construction (2026).pdf 9. "AIIA Network Operations and Maintenance Agent Evaluation Standard"

Achievement introduction: It specifies the evaluation methods for network operations and maintenance agents, covering the evaluation framework, testing process, scoring rules, and grade determination criteria. It aims to provide providers with a basis for self-testing product capabilities, provide application parties with a reference for selection, promote the standardized and systematic development of the technical capabilities of network operations and maintenance agents, and facilitate the continuous evolution of information and communications network operations and maintenance toward full-scenario autonomy. 10. "Device-Side Intelligence Industry Development Report (2026)"

Achievement introduction: The report systematically reviews the development context, policy environment, market situation, and conceptual characteristics of the device-side intelligence industry, with a focus on key technologies such as model lightweighting, heterogeneous hardware integration, foundational software, agents, device-cloud collaboration, and security and trustworthiness; studies native product forms such as personal intelligence, spatial intelligence, mobile intelligence, industry intelligence nodes, and embodied intelligence; reviews the development characteristics of consumer entertainment, smart home, smart office, smart travel, smart health, and integrated scenarios; and further assesses the development trends, main challenges, and advancement paths of device-side intelligence in technology, applications, ecosystem, and security, providing reference for technological innovation, industrial layout, standard development, and application promotion. Download link: Device-Side Intelligence Industry Development Report (2026).pdf 11. "Device-Side Intelligence Testing Platform Co-Construction Plan" and "Device-Side Intelligence Testing Platform 1.0"

Achievement introduction: The China Academy of Information and Communications Technology, together with Huawei, ZTE, vivo, Honor, Xiaomi, China Unicom, and other units, jointly launched the "Device-Side Intelligence Testing Platform Co-Construction Plan" and officially released "Device-Side Intelligence Testing Platform 1.0." The platform targets multiple types of intelligent terminals and, focusing on key directions such as device-side agent capabilities, model capabilities, and system components, builds a testing and verification system covering test cases, datasets, real-device execution, result determination, and report generation. Around different terminal forms and application scenarios, the China Academy of Information and Communications Technology, together with industry partners, launched the "Device-Side Intelligence Testing Platform Co-Construction Plan" to jointly advance the development of testing standards, cases, datasets, evaluation methods, and tools. 12. Device-Side Intelligence Series Standards

Achievement introduction: The series standards "Device-Side Intelligence—Capability Grading Requirements for Artificial Intelligence Operating Systems (AIOS)," "Device-Side Intelligence—Performance Testing Methods for Device-Side Models," "Device-Side Intelligence—Technical Requirements for Intelligent Assistants (Claw)," "Device-Side Intelligence—Technical Requirements for Multimodal Interaction," and "Device-Side Intelligence—Technical Requirements for Agent Payments." 13. "AI Native Research Report 1.0—Enterprise AI Native Transformation Framework and Practice"

Achievement introduction: The report defines AI Native as "products, enterprises, or systems built with artificial intelligence as the core driving force from the outset of design," with the criterion for judgment being whether the operating logic still holds after AI is removed; it possesses four major characteristics: AI as the core, an efficient data flywheel, close human-machine collaboration, and continuous evolution. The report proposes an enterprise transformation capability framework covering six dimensions—strategy and culture, organization and governance, technology and platforms, data and knowledge, business and applications, and security and trustworthiness—and provides action recommendations in seven areas: scenario identification, organizational and governance restructuring, technology systems, data and knowledge, digital employees, security governance, and frontline deployment engineering. In terms of trends, the report points out that artificial intelligence is shifting from an optional tool invoked on demand to a fundamental operating condition for enterprises, with the center of value shifting from the model layer to the semantic layer, and the focus of competition lying in business process restructuring and the depth of domain knowledge accumulation. 14. "AI Native Research and Evaluation Testing System"

Achievement introduction: The "AI Native" series of evaluation tests is a third-party evaluation and testing system launched by the China Academy of Information and Communications Technology (CAICT). This system is grounded in the core connotation of "AI Native"—that is, taking artificial intelligence as the core hub from the outset of design and systematically restructuring the technical architecture and business logic. It aims to comprehensively evaluate the level of enterprises' AI Native capability building, measure technical product capabilities and scenario practice results, and help enterprises leverage the AI Native paradigm to activate data potential and achieve a fundamental upgrade of production and operation models. This system is built on CAICT's more than ten years of accumulated experience in trusted digital and intelligent evaluation and testing, cuts in from both the supply side and the application side, and covers core directions such as overall capabilities, data and knowledge, technology and engineering, scenario applications, and security assurance. 15. Digital Employee Series Standards

Achievement introduction: "Digital Employee Operations Management Capability Maturity Assessment Model" is aimed at enterprise organizations, evaluates the maturity of operations management capabilities, and sets three levels: Basic, Excellent, and Outstanding; "Digital Employee Capability Grading Requirements" is aimed at individual digital employees, evaluates capabilities from six major dimensions, and sets five levels from L1 to L5. The two serve as inputs and outputs for each other, forming an organic whole of "managing the organization and evaluating the individual," and providing a unified yardstick for full-lifecycle management. The release of the two standards marks that the digital employee industry has entered a new stage of standardized development and provides support for the deep integration of artificial intelligence and the real economy. 16. "Research Report on the Construction Path and Industry Practice of Enterprise Computing Power Foundations in the Digital-Intelligent Era"

Achievement introduction: "Research Report on the Construction Path and Industry Practice of Enterprise Computing Power Foundations in the Digital-Intelligent Era" was jointly compiled by Hygon Information Technology Co., Ltd., the Institute of Artificial Intelligence of the China Academy of Information and Communications Technology, the Hygon Industrial Ecosystem Cooperation Organization, and the Basic Software, Hardware and Ecosystem Working Group of the Artificial Intelligence Industry Alliance of China. Against the backdrop of an era in which artificial intelligence technologies represented by large models are moving from the exploration stage to large-scale implementation and computing power has surged to become a core factor of production driving industrial upgrading, the report focuses on the practical challenges commonly faced by industry users in planning and implementing computing power foundations, such as high cognitive thresholds, limited construction experience, and a lack of reference cases. By systematically sorting out enterprise computing power foundation construction paths, methodologies, and practice frameworks, and deeply mining mature construction cases in industries such as manufacturing and energy, it provides practical demonstrations and decision-making references that can be consulted and reused by various enterprises currently exploring computing power construction, helping the industry accelerate toward a new stage of intelligent development. Download link: Research Report on the Construction Path and Industry Practice of Enterprise Computing Power Foundations in the Digital-Intelligent Era.pdf 17. "Token Efficiency: The Key Variable of Green Computing in the AI Era"

Achievement introduction: The research report "Token Efficiency: The Key Variable of Green Computing in the AI Era" was jointly released by Ant Group and the Institute of Artificial Intelligence of the China Academy of Information and Communications Technology. Taking token efficiency as the key entry point, the report systematically analyzes the new challenges of green computing in the AI era, constructs a full-stack technical system framework for token efficiency from the model layer to the application layer, systematically expounds the connotation of token efficiency, and, combined with typical application practices, proposes a future-oriented token efficiency measurement system. Download link: Token Efficiency: The Key Variable of Green Computing in the AI Era.pdf 18. "Guidelines for Building High-Quality Datasets for Agent Capability Construction (2026)"

Achievement introduction: "Guidelines for Building High-Quality Datasets for Agent Capability Construction (2026)" was launched for compilation in April 2026, jointly compiled by the Institute of Artificial Intelligence of CAICT and Lenovo (Beijing) Co., Ltd., and received active participation and support from more than 30 units including Neusoft. The guidelines focus on the new requirements that agent capability construction places on high-quality datasets, systematically summarize dataset construction ideas and practical experience, and provide useful references for the development of the agent industry and the construction of high-quality datasets. Download link: Guidelines for Building High-Quality Datasets for Agent Capability Construction (2026).pdf 19. "AIIA Technical Specification: Token Security Protection Capability Requirements"

Achievement introduction: To clarify the capability requirements and evaluation basis for token security protection in generative artificial intelligence scenarios, and to standardize the planning and design, R&D and testing, deployment and implementation, operations management, and security assessment of related products and security components, the China Academy of Information and Communications Technology (CAICT), together with China Mobile Communications Group Co., Ltd., China Mobile (Suzhou) Software Technology Co., Ltd., China Telecom Cloud Technology Co., Ltd., Beijing Volcano Engine Technology Co., Ltd., Huawei Technologies Co., Ltd., Beijing Topsec Network Security Technology Co., Ltd., Beijing Xin'an Century Technology Co., Ltd., Arm Technology (China) Co., Ltd., Venustech Information Technology Group Co., Ltd., Deyi Ziji (Fujian) Digital Technology Co., Ltd., and AsiaInfo Technologies (Chengdu) Co., Ltd., released this technical specification. 20. "AIIA Technical Specification: AI SOC Capability Requirements and Evaluation Methods"

Achievement introduction: CAICT, together with China Mobile, NSFOCUS, Qi An Xin, Topsec, and many other units, jointly compiled the "AI SOC Capability Requirements and Evaluation Methods" AIIA technical specification and officially released it here. The specification defines the overall framework of an artificial intelligence-enabled Security Operations Center (AI SOC), as well as general technical requirements in areas such as data and knowledge, artificial intelligence capabilities, security operations, and system management. It is applicable to the planning, construction, operation, and improvement of artificial intelligence-enabled Security Operations Centers, and provides a compliance reference for intelligent security operations centers.

Source: Artificial Intelligence Industry Alliance (AIIA) www.smartcity.team

Disclaimer: Please indicate when reprinting that the source is "Smart City Industry Analysis (www.smartcity.team)" and the relevant author. Except for original content, some information on this site comes from the Internet. If the relevant content involves infringement of rights, please contact the contact information at the bottom to delete it.

[Digital Technology Special Research] Community

In-depth peer exchange, focused on smart city and digital technology industry research and material accumulation.

Continuously updating the latest industry white papers, industry research, policy developments, and implementation solutions to help industry practitioners efficiently accumulate knowledge and share experience.

Massive premium content is continuously iterated, providing cutting-edge perspectives, connecting industry talent, carrying out case co-creation and resource exchange, and jointly building a high-quality industry exchange atmosphere.

#post-##

Recommended for you

2026 Embodied Intelligence Industry Innovation Report: Zhiyuan Innovation Day, jointly released by Xinhua News Agency and a CAICT subsidiary, defining industry stages with the "XYZ curve" of embodied intelligence industry development, driving technological evolution with "three intelligences in one," and measuring productivity value with Worken

Jiazi Guangnian Think Tank "2026 Agent Commercialization and Competition Path Research Report": Analysis from five aspects: product and demand evolution, formation of commercial value, market ecosystem and enterprise deployment, application landscape and commercialization practice, and industry evolution and competition outlook

"Embodied Intelligence Development Report (2026)": Embodied intelligence is at a transitional node toward pilot applications; the industry is revealing coldness amid continued heating up; the technical system composed of the four elements of models, data, bodies, and computing power is evolving collaboratively; industrial chain division of labor is becoming more refined, but a sustainable commercial closed loop still needs to be established

National Public Data Operations Development Report (2025-2026): There are 76 public data resource authorized operation platforms or zones that have been publicly disclosed as operating or under construction nationwide, with 202 determined implementing institutions and 224 operating institutions

Embodied Data Development Report 2026: Detailed explanation of the current development status, meaning and classification, industrial chain and business models, distinctive cases, and development trends of the embodied data industry

Device-side Intelligence Industry Development Report (2026): A systematic review of the device-side intelligence industry overview, technical system, product forms, and application scenarios