Guangdong Province releases several measures on artificial intelligence empowering thousands of industries

Guangdong Province releases several measures on artificial intelligence empowering thousands of industries

Invest in Zhaoqing 2024-06-07 20:13 Guangdong

On June 6, the General Office of the People's Government of Guangdong Province issued the "Several Measures of Guangdong Province on Artificial Intelligence Empowering Thousands of Industries". The full text is as follows:

Several Measures of Guangdong Province on Artificial Intelligence Empowering Thousands of Industries

In order to implement the decisions and arrangements of the Party Central Committee and the State Council on promoting the development of artificial intelligence, promote the high-quality development of the artificial intelligence industry in our province, accelerate the formation of new quality productivity, build a modern industrial system, empower thousands of industries to improve quality and efficiency, and create a new economic model, new life experience, and new governance methods in the intelligent era, these measures are formulated.

I. Overall goal

By 2025, the province's computing power will exceed 40EFLOPS, and the scale of the core artificial intelligence industry will exceed 300 billion yuan. By 2027, the province's artificial intelligence industry foundation will be further consolidated, with a computing power scale of more than 60EFLOPS, and the country's leading algorithm system and computing power network system will be basically formed; the supply of intelligent terminal products will be abundant, and more than 100 large-scale intelligent terminal products will be created in eight categories such as mobile phones, computers, homes, and robots, and the scale of the core artificial intelligence industry will exceed 440 billion yuan; focusing on manufacturing, education, and elderly care, more than 500 application scenarios will be created, and the labor productivity of all walks of life will be significantly improved.

2. Consolidate the foundation of the artificial intelligence industry and accelerate the formation of new quality productivity

(I) Increase the supply of artificial intelligence core chip devices.

1. Establish an artificial intelligence chip ecosystem. Build an adaptive chip development ecosystem, and increase the development and production of high-performance, low-power end-side chips for home appliances, security monitoring, medical equipment, etc. Encourage enterprises to promote the research and development of communication, display, audio and other modules through integrated processors, radio frequency communications, smart sensors, memory, etc. Cultivate a chip innovation and development ecosystem, explore the research and development and application of chips such as storage and computing integration, brain-like computing, core particles, and instruction sets, promote chip applications for the cloud and terminals, and promote high-performance cloud-based intelligent servers. By 2027, the AI ​​chip ecosystem will be initially established.

2. Create an intelligent perception industry system. Build intelligent sensor industry clusters and characteristic industrial parks, promote the development and industrialization of sensors such as image, sound, and touch, and accelerate the development and large-scale production of sensors such as biometric recognition and image perception in consumer electronics, home appliances, and other fields. Promote the synergy of industrial ecosystems such as processing and manufacturing, integrated packaging, and metrology and testing. By 2027, the scale of the high-end intelligent sensor industry will double.

(II) Promote the iteration and upgrade of AI software.

3. Strengthen the R&D and innovation of intelligent software. Develop an engine framework tool system, and develop a high-performance distributed parallel computing framework and platform for inference acceleration engines and ultra-large-scale deep learning. Develop intelligent operating systems, and accelerate the development of server-level intelligent operating systems and intelligent terminal operating systems with independent intellectual property rights. Encourage enterprises to deeply explore the needs of key industries such as manufacturing, medical care, and education, strengthen the mutual adaptation, performance optimization, and application promotion of AI framework software and hardware, and create an integrated software and hardware ecosystem. By 2027, the scale of the intelligent software industry will reach 27 billion yuan.

4. Broaden the breadth of intelligent software applications. Implement the industrial software efficiency enhancement project, and use artificial intelligence to improve the R&D efficiency of the underlying technologies of industrial software such as 3D modeling, parametric design, and entity segmentation. Implement the application software quality improvement project, integrate the scene data, technology, and process of various industries, and provide professional industry software solutions. Implement the terminal software promotion project to support software companies to develop various commercial APPs, office software and other terminal software for artificial intelligence. By 2027, the coverage rate of artificial intelligence independent software will reach 50%.

(III) Systematically build an algorithm industry matrix.

5. Build a high-quality Chinese data set. Strengthen the integration of data resources in government information systems, and build a public and public trustworthy data space and public data set based on the government network. Support local areas to carry out public data operation exploration based on artificial intelligence capabilities, select advantageous industrial fields, and promote the efficient integration and utilization of public data and social data. Encourage enterprises to build high-quality Chinese corpus databases for industries, and promote the collection, sharing and use of typical industry data. By 2027, build more than 50 high-quality industry data sets.

6. Accelerate the supply of algorithm products. Develop professional algorithm products and tools around the domestic mainstream artificial intelligence development framework. Encourage enterprises to increase algorithm applications and create a batch of algorithm products and application demonstration projects. Accelerate the development and iteration of general large models and vertical industry large models, make detailed scene-specific models, and support lightweight, efficient, and easy-to-deploy small and medium-sized models. Support the development of independent and controllable large model products, and publish product and service catalogs. By 2027, create 100 algorithm products with demonstration and promotion effects.

7. Build a large model open source community. Build an original basic large model resource pool, encourage the establishment of a large model open source community, and support the free and open access, parameter adjustment, and application development of large models and their derivatives. Optimize the development environment of large models and reduce non-technical factors. Carry out training, knowledge popularization and other activities for community members to cultivate large model development talents. By 2027, basically build a one-stop open source service system for research, development, collaboration, deployment and on-site display of artificial intelligence.

(IV) Moderately advance the deployment of computing power network construction.

8. Accelerate the construction of intelligent computing power infrastructure. Issue and implement policies and measures to accelerate the development of computing power infrastructure in our province, and promote efficient complementarity and coordinated linkage between the Pearl River Delta and eastern, western and northern Guangdong. Accelerate the construction of national hub node data centers, and strengthen the city edge intelligent computing center around key application scenarios. Strengthen government-enterprise cooperation, accelerate the collection of existing training computing power, and enhance shareable computing power. By 2027, the intelligent computing coverage of key industries will be significantly improved.

9. Strengthen the intensive supply of computing power. Promote the coordinated development of intelligent computing power and general computing power such as operators, cloud service providers and various computing power platforms to meet the computing power needs of various businesses such as balanced, computing and storage intensive. Accelerate the construction of computing power public platforms such as Guangzhou Artificial Intelligence Public Computing Center, Pengcheng Cloud Brain, and Hengqin Advanced Intelligent Computing Center, and support the research and development of Pengcheng Laboratory's "China Computing Network". Explore the construction of a multi-level computing power elastic scheduling platform, integrate computing power resources in a cloud service manner, and realize cross-domain scheduling and orchestration of multi-heterogeneous computing power. By 2027, basically realize efficient sharing, autonomous collaboration and unified services of computing power resources.

10. Promote the optimization of computing power network layout. Support the research and development of optical communication products and equipment, and build a 400G/800G all-optical high-speed low-latency capacity network. Promote the deployment of metropolitan optical transmission equipment to integrated access nodes and user sides, and the latency between important computing infrastructure in urban areas shall not exceed 1ms. Support the transformation of optical fiber links in key places to improve the service and dispatching capabilities of the transportation network. By 2027, the province's high-bandwidth, low-latency computing network support capabilities will be greatly improved.

III. Build a new highland for intelligent terminal products and shape a new image of Guangdong brand

(I) Improve intelligence and strengthen high-end equipment.

11. Promote the innovative development of intelligent robots. Accelerate the research and development and production of products such as machine brains, machine limbs, machine bodies, and general products, and promote the development and application of embodied intelligent robots such as humanoid robots. Develop high-precision industrial robots such as collaborative robots and adaptive robots with flexible interaction and dynamic path planning, and accelerate the application and promotion of service robots with intelligent human-machine interaction and multi-degree-of-freedom precision control. By 2027, the operating income of the intelligent robot industry will reach 90 billion yuan.

12. Accelerate the intelligent upgrading of high-end equipment. Promote the intelligent transformation of high-end equipment such as high-end CNC machine tools, marine engineering equipment, aviation equipment, satellites and applications, and rail transit equipment. Support the development of highly autonomous, highly airworthy, and highly reliable unmanned boats and unmanned submersibles. Strengthen the development and industrialization of equipment in the fields of deep-earth resource exploration, urban underground space development and utilization, polar exploration and operations. By 2027, the operating income of the high-end equipment manufacturing industry will reach 380 billion yuan.

13. Intelligent networking supports new energy vehicles. Create intelligent networked vehicle terminals that integrate safe travel, smart life, mobile office and other functions. Develop intelligent networked terminals such as unmanned taxis, smart buses, and smart heavy trucks, and encourage intelligent, one-stop transportation and travel service applications. By 2027, the annual output of new energy vehicles will reach more than 3.5 million.

(II) Empowering intelligence and strengthening consumer terminals.

14. Innovating operating systems to create artificial intelligence mobile phones. Support the deep integration of efficient and low-loss lightweight large models in operating systems, and promote the application and innovation of terminal-side model algorithms and architectures. Support enterprises to combine multimodal natural interaction technology on AI OS, enhance real-time interaction capabilities, expand mobile phone application scenarios, and create personal digital assistants. By 2027, the output of AI mobile phones will reach more than 100 million.

15. Deploy exclusive models to build AI computers. Use AI technology to continuously learn users' personal data and habits, form a personal local knowledge base that does not go out of the end, does not go to the cloud, is transferable, inheritable, and can grow, and combine lightweight technologies such as model compression to form personal exclusive models and new intelligent production tools. By 2027, the output of AI computers will reach more than 30 million.

16. Focus on "old and young" to strengthen intelligent childcare and rehabilitation products. Improve the intelligence level and safety of intelligent daily auxiliary products such as home service robots, health monitoring equipment, wheelchairs, rehabilitation training and health promotion aids such as walkers and incontinence training aids, and elderly care products such as intelligent bathing aids, transfer machines, and rehabilitation nursing beds. Support the research and development of intelligent interactive toy products that integrate chat, early education enlightenment, and entertainment companionship. By 2027, more than 1,500 new professional products will be added in the field of childcare and rehabilitation products.

17. Integrate cross-border and improve the quality of smart home appliances. Deeply integrate learning algorithms, image recognition, intelligent voice and other new technologies into home appliances, develop high-end new intelligent products, and enhance the international influence of brands. Promote the interconnection of smart home products such as home appliances, furniture, and daily necessities, develop personalized customization, special-purpose and special-purpose smart home appliances, and enrich the supply of high-end smart home appliances and services. By 2027, the operating income of the strategic industrial cluster of smart home appliances will reach 1.9 trillion yuan.

18. Enrich virtual reality smart terminals. Deepen the integration and innovation of artificial intelligence technology with key virtual reality technologies such as near-eye display, rendering processing, perception interaction, network transmission, content production, compression coding, security and trustworthiness, develop and produce a batch of integrated and split terminal products, and make breakthroughs in key areas such as industrial production, cultural tourism, integrated media, education and training, sports and health, business and trade creativity, and smart cities. By 2027, more than 3,000 new virtual reality terminals and applications will be added.

IV. Create a new engine for intelligent integration applications and form a new outlet for economic growth

(I) Empower new momentum for the real economy.

19. Promote the digital and intelligent transformation of industry. Promote the intelligentization of the entire manufacturing process by "replacing people with machines and replacing brains with data", and deepen the integrated application of artificial intelligence technology in R&D design, pilot verification, production and manufacturing, and operation management. Accelerate the empowerment of key industries such as raw materials, equipment manufacturing, consumer goods, and electronic information. Build a large model for carbon emission management, accelerate the green transformation of production, strengthen energy conservation and carbon reduction in key high-energy-consuming industries such as steel, non-ferrous metals, and building materials, and reach 700 national-level green factories.

20. Deep integration promotes intelligent construction. Integrate building information modeling (BIM), digital twins and other technologies with multimodal large models as carriers, promote full-professional information integration and design collaboration, realize automatic completion of design analysis, consistency review of construction drawings and building models, data matching and other tasks, shorten the conversion time of design models to construction applications, and improve the intelligence level of project management, quality and safety supervision and other links.

21. Intelligent systems drive smart agriculture. Accelerate the application of intelligent facilities, use intelligent sensors to monitor soil moisture, temperature and other parameters, strengthen the application of artificial intelligence algorithms in yield judgment, weather forecasting, market analysis and other aspects, and build digital pastoral and smart farms (animal husbandry and fishery). Improve the level of intelligence in animal and plant protection and strengthen pest and disease control. Accelerate the manufacturing and promotion of "one large and one small" intelligent agricultural machinery equipment, and encourage new mechanisms and models of agricultural machinery industry chain and supply chain.

22. Smart transportation creates a modern transportation system. Carry out demonstration projects such as intelligent railways, highways, waterways, ports, etc., promote smart hubs, postal services, and maritime affairs, and build a smart operation system for the core road network in the Greater Bay Area. Promote the integrated application of "vehicle-road-cloud" for intelligent connected vehicles, build digital intelligent infrastructure, city-level service management platforms, and explore the safe application of high-precision maps. Efficiently promote the construction of low-altitude economic infrastructure and accelerate the construction of a production and supply system. Develop new logistics infrastructure such as smart logistics parks and digital warehouses, and encourage cities with conditions to build a smart logistics "brain".

23. Multi-party integration and co-construction of smart energy. Accelerate the integration and application of artificial intelligence with equipment and systems such as power, energy industrial Internet, and power global Internet of Things. Promote the construction of smart substations and multi-energy complementary integrated energy networks, and build a province-wide smart power grid that adapts to large-scale new energy access and meets the "plug and play" requirements of distributed energy.

(II) Empower new experiences in smart people's livelihood.

24. Intelligent assistance improves the quality of education and teaching. Promote the development and application of new resources such as intelligent teaching assistants, intelligent guidance, and educational robots, and help reduce burdens and increase efficiency through human-machine co-teaching and human-machine co-education, and innovate teaching models. Develop intelligent evaluation tools and promote the reform of education and teaching evaluation methods. Strengthen the equipment of virtual simulation, intelligent perception, etc., provide adaptive learning resources and intelligent learning services, and create smart classrooms and smart campuses.

25. Interconnection and sharing improve medical service capabilities. Build smart hospitals and integrate and connect relevant online service terminals. Establish a cross-departmental and cross-institutional public health data sharing and scheduling mechanism and a smart early warning multi-point trigger mechanism, and strengthen smart monitoring of drug supply guarantee. Create "smart medical assistants" to assist doctors in improving diagnostic efficiency and quality, assist patients in self-checking symptoms, guiding medication, interpreting reports, etc. Strengthen the application of robots in patient pre-hospital management, in-hospital diagnosis and treatment, and post-hospital rehabilitation tracking service systems.

26. Smart elderly care promotes the silver economy. Strengthen the configuration, integration and effective connection of health and elderly care resources for individuals, families, communities and medical institutions, and create a smart elderly care service system of "platform + service + product + response + supervision". Promote the intelligent transformation of commonly used Internet applications, APP applications, mobile terminals, and home appliances for the elderly in the fields of medical care, civil affairs, and living payment.

27. Smart customization promotes national fitness. Support multimodal big models to integrate professional knowledge in the fields of physiology, diet, fitness equipment, etc. to create private health coaches. Promote the deep integration of national fitness and artificial intelligence, develop smart sports products, build smart sports venues and smart outdoor sports facilities, and promote new smart sports technologies, new products and application scenarios such as sports fitness apps and platforms.

28. Smart housekeeping creates life assistants. Promote comprehensive housekeeping robots that integrate cleaning, education, catering, shopping, nursing, entertainment and other functions. By extracting and analyzing deep information of human language and generating action instructions, complete the machine replacement of family affairs and provide personalized services.

(III) Empower new social governance efficiency.

29. Benefit enterprises and the people to build a smart government. Use the intelligent government big model to upgrade the Guangdong Government Service Network and the "Guangdong Series" government service platform to provide full-time online question and answer and search services. Build a video computing support basic platform for party and government agencies at all levels across the province to improve the level of video intelligence. Integrate economic algorithm models in the "Guangdong Economy" platform to provide decision-making support for the government in scenarios such as economic operation monitoring, analysis and judgment, and policy simulation.

30. Smart construction releases new vitality in cities. Promote the application of technologies in the fields of community governance, natural resources, market supervision, etc., expand the application of scenarios such as digital-real integration of urban information model platforms, and realize the monitoring, early warning and prevention of ecological risks and urban operation safety risks. Promote the intelligent access to the bottom line of people's livelihood, and support artificial intelligence products and services that protect the rights and interests of special groups such as the elderly, the weak, the sick and the disabled.

31. Full-process assistance improves judicial efficiency. Promote artificial intelligence technology to provide full-process high-level auxiliary support for trial execution, litigation services and other work, realize the automatic generation of evidence guidance and review records, case judgment deviation warning, automatic archiving of electronic files, litigation and mediation consultation and Q&A and other smart functions, and efficiently ensure clean justice.

32. Smart security improves security capabilities. Promote the construction of smart food and drug supervision and strengthen smart safety supervision of food and drugs. Deepen the application of smart prevention and control, perception and other technologies in the field of public security, and intelligently warn of natural disasters. Promote the use of smart robots, drones and other technologies for rapid emergency response. Promote the intelligent transformation of public safety facilities such as gas, bridges, underground pipelines, and mines, fire protection and other fields to ensure safe production.

33. Smart environmental protection creates a better home. Build a multi-faceted intelligent environmental governance system to comprehensively improve the capabilities of pollution source tracing, pollution prevention and control, situation analysis, and decision-making support. Improve the ecological environment monitoring network, and promote the intelligence of remote sensing monitoring, carbon monitoring, new pollutant monitoring, and environmental monitoring and assessment. Promote the intelligence of ecological environment supervision and law enforcement.

(IV) Empower new digital consumption formats.

34. Smart business stimulates business potential. Promote new business services such as precision marketing, smart business districts, and smart distribution, develop "unmanned stores", and promote the intelligent transformation of business circulation services. Based on segmented data such as consumer historical purchasing behavior and transaction records, intelligently identify consumer's significant characteristics, generate user portraits, predict preferences and interests, and enhance user experience through personalized marketing.

35. Smart finance improves financial services. Integrate intelligent models throughout all processes of product services, and use technologies such as robotic process automation, natural language processing, and intelligent character recognition to reconstruct digital processes end to end. Build a multi-faceted service channel and promote the upgrade of physical outlets to multi-modal, interactive smart outlets. Focus on daily high-frequency financial scenarios for the elderly, the disabled, and ethnic minorities, and create an intelligent service system.

36. Innovate experience to upgrade the tourism development model. Build smart scenic spots, and develop smart guided tours, VR/AR immersive tourism, cloud tourism and other application scenarios on a large scale. Expand the application capabilities of large-scale model technology in tourist attractions in passenger flow statistics, consumption analysis and prediction, emergency response, etc., and expand the marketing channels of tourism products.

37. Intelligent interaction enriches cultural and entertainment life. Build smart libraries and museums, and create information digital service platforms such as smart radio and television and film digital program management. Encourage enterprises to use intelligent technology to transform and upgrade the industrial chain, create intelligent-driven digital people, and promote the modernization of content production and dissemination methods such as cultural and life videos. Cultivate immersive interactive formats and build a batch of new scenarios and new applications.

(V) Empower new fields in all walks of life.

38. Comprehensively promote the application of smart scenarios. Actively play the role of artificial intelligence in scientific research fields such as drug design and discovery, materials science, physical and chemical simulation, and promote scientific progress and innovation. Accelerate the diversified application of artificial intelligence in various fields such as biomanufacturing, commercial aviation, catering, real estate, etc., accelerate the formation of new consumption scenarios, new formats, new models, and new brands, further liberate the labor force, and significantly increase new jobs related to artificial intelligence applications.

V. Guarantee measures

39. Establish and improve the organizational coordination mechanism for the development of the artificial intelligence industry. Under the overall guidance of the Provincial Party Committee's Science and Technology Committee, the Provincial Leading Group for High-quality Development of the Manufacturing Industry will strengthen the organization and coordination of artificial intelligence empowering thousands of industries, give full play to the role of the Expert Advisory Committee, and study and solve the difficulties and problems in industrial development. Establish a provincial-municipal linkage and government-enterprise collaboration mechanism, and encourage cities with conditions to include the artificial intelligence industry in the key areas of strategic emerging industries in the city. Support the establishment of an inclusive and prudent fault-tolerant and error-correcting mechanism, and organize various market entities to implement partner cooperation plans.

40. Multi-party participation to strengthen policy supply. Coordinate various provincial and municipal funds, strengthen financial support for the artificial intelligence industry, and support project construction in the field of artificial intelligence. Study and promote the formulation of laws and regulations related to the supply of trusted artificial intelligence products and services. Regularly publish a list of application scenarios, establish an incentive mechanism for innovative products, and compile a recommended catalog of innovative products. Encourage cities to provide relevant policy support for the first set of equipment, the first version of software, the first batch of new materials, and the first purchase order, and deploy and use them in application scenario units.

41. Gradually establish an industrial standard system. Build a patent database in the field of key technologies in artificial intelligence, carry out patent navigation, form a patent alliance in subdivided fields, and promote the transfer and transformation of intellectual property rights. Accelerate the layout of standards, specifications and tools for large-scale model safety and compliance certification, lightweight technology, performance evaluation, etc., and support enterprises to participate in the formulation. Establish institutional norms to ensure the healthy and safe development of artificial intelligence. Explore the establishment of an artificial intelligence industry empowerment indicator system.

42. Cultivate innovative talents. Increase the efforts of various policies and measures to attract and gather talents, focus on cultivating and gathering high-level talents, and drive the cultivation of a group of artificial intelligence innovation teams. Build bases for talent continuing education and training and high-skilled talent training, and carry out training for urgently needed professional and technical talents and high-skilled compound talents in artificial intelligence. Implement the "Smart Craftsman" and "Chief Algorithmist" plans in advantageous industries, explore the classification and statistics of artificial intelligence talents, and regularly publish talent industry standards.

43. Strengthen innovative demonstration applications. Support a number of key projects with high innovation, great application value and good market prospects, forge artificial intelligence "long board" enterprises, summarize demonstration results and excellent cases in a timely manner, and give priority to promotion opportunities. Organize and carry out artificial intelligence innovation and entrepreneurship competitions, showcase innovative products, excellent cases and application scenarios, carry out investment promotion and project promotion and other docking services, and build a supply and demand platform for new products and new technologies.

44. Create an industrial ecological atmosphere. Give full play to the role of Guangzhou and Shenzhen in building the National New Generation Artificial Intelligence Innovation Development Pilot Zone and the National Artificial Intelligence Innovation Application Pilot Zone, and accelerate the construction of the national (provincial) artificial intelligence open innovation platform. Encourage local cities to layout key industrial clusters according to local conditions, and build a number of industrial bases such as artificial intelligence characteristic towns and intelligent terminal characteristic industrial parks. Intelligently upgrade the industrial supporting services and operation systems of the park to create a smart park. Support the construction of artificial intelligence industry innovation centers, empowerment centers, cloud computing service platforms, etc., and provide one-stop ecological services.

45. Strengthen international exchanges and cooperation. Encourage foreign-invested enterprises to set up artificial intelligence laboratories and R&D centers in Guangdong. Actively expand the international market, encourage the transformation of offshore innovation results in China, and treat them as domestic innovation results. Encourage original design manufacturers to do a good job in international development and enhance the design capabilities of intelligent products. Deepen domestic and foreign cooperation and exchanges, improve the multi-level communication and cooperation mechanisms of international organizations, industrial alliances, well-known enterprises, and organize and introduce international high-level artificial intelligence academic conferences.