On the afternoon of December 5, the 2025 Digital Transformation Conference of the Textile and Apparel Industry was held in Jiujiang City, Jiangxi Province. At the meeting, Sun Ruijie, Secretary of the Party Committee and President of the China Textile Industry Federation, delivered a speech on "Artificial Intelligence Drives New Quality Productivity in the Textile and Apparel Industry". The full text of the speech is as follows:
The Fourth Plenary Session of the 20th CPC Central Committee has outlined a blueprint for the future. The session has placed "building a modernized industrial system and consolidating and expanding the foundation of the real economy" at the top of strategic tasks, emphasizing the need to "promote the deep integration of the real economy and the digital economy", "fully implement the action of 'AI+'", and "seize the commanding heights of the application of the AI industry and empower thousands of industries in an all-round way". This provides fundamental guidance and important instructions for the construction of a modernized industrial system for the textile industry in the new era.
At present, digital transformation is no longer a mere technological iteration, but a paradigm revolution that touches upon the mode of production, industrial forms, and value models. Looking at the present and facing the future, this conference focuses on the issue of digital transformation, through the complex technical phenomena, to explore the industrial changes and opportunities brought by AI. This is not only a positive practice to implement the spirit of the Fourth Plenary Session of the Nineteenth Central Committee, but also a useful discussion to find the path of industrial upgrading at a key node. The holding of the conference is of great significance. On this occasion, I would like to share a few points of understanding.
1, To understand the significant importance of digital transformation for the development of the textile industry.Around the requirements for high-quality development, the textile and apparel industry is making a leap towards "innovation-driven technological industries, culture-led fashion industries, responsibility-oriented green industries, and people-oriented health industries". In this process, digital transformation is not a simple technological upgrade, but a strategic lever and core variable for building a modern industrial system.
First, reshape the power mechanism to drive the systematic upgrade from "element-driven" to "innovation-emergent". The digital economy constitutes a brand-new innovation ecology, where digitalization, as a "neural network", connects material innovation, process innovation, and manufacturing innovation through the deep integration of technology, data, and business. It not only breaks the information barriers in the industrial chain but also evolves the logic of industrial innovation from a single technological breakthrough to systematic innovation along the entire chain.
Second, the reconstruction of the value form drives the paradigm shift from "material production" to "cultural leadership". Through the empowerment of data elements, digitalization transforms invisible cultural values into computable and perceptible productivity. From the source innovation of AI-assisted design to the scenario construction of virtual reality, digitalization extends the focus of the industry from products to services, building a value ecology from the real to the virtual.
Third, transform the governance model to achieve a fundamental shift from "coarse growth" to "lean governance". By optimizing process parameters and management systems with intelligent algorithms, digital technology significantly enhances resource efficiency from the source and curbs the generation of pollutants. The transparent supply chain system built with blockchain not only provides a reliable foundation for carbon footprint tracking and green certification but also serves as a key support for green finance and trade. This has enabled the industry to establish a new pattern of lean governance that advances "carbon reduction, pollution control, green expansion, and growth" in a coordinated manner.
Fourth, sublimate the service attributes and complete the transition from "supplying products" to "ensuring health". The textile industry is deeply integrating into the big health scenario, building a value network that guarantees the health of life, life, and occupational health, and realizing the return of industrial value to serving the people and humanistic care. For example, through smart wearable devices to achieve real-time perception of human health, through smart transformation to optimize the working environment, etc.
2, it is necessary to grasp the realistic foundation and characteristic trends of the digital transformation of the textile industry.The textile industry is one of the earliest and most fully developed industries in the practice of the integration of two modernizations. As of September 2025, the level of the integration of two modernizations in the industry has reached 64.8. The coverage of the application of digital technology is continuously increasing. As of September 2025, the proportion of enterprises in the textile and apparel industry that have been fully digitalized in key business links such as research and development and design, production and manufacturing, and operation and management has reached 64.9%, which is higher than the national manufacturing industry average of 61.3%; The rate of key processes under numerical control has reached 65.3%, which is higher than the national consumer goods industry average of 64.4%. The application popularity of industrial software has steadily increased. As of September 2025, the penetration rates of digital R&D and design tools, ERP, PLM, and MES in textile enterprises have reached 84.9%, 71.1%, 30.2%, and 32.3%, respectively, all of which have achieved rapid growth; The digitalization rate of production equipment and the network connection rate of digital equipment in textile enterprises have reached 59.1% and 53.5%, respectively, which are higher than the national manufacturing industry averages of 57.9% and 52.6%. The proportion of textile industry enterprises that are networked and collaborative, and the penetration rate of industrial e-commerce have reached 51.3% and 72.2%, respectively, which are higher than the national manufacturing industry averages of 47.2% and 71.9%. The level of intelligent and networked transformation of textile enterprises is continuously improving. The results of the digital transformation of the textile industry are fruitful and it is at the forefront.
The first is the transformation from standard production to flexible manufacturing. Driven by data, production meets the needs of "small batch, multiple varieties, personalization, and customization" through intelligent manufacturing, which has become an important trend. With the construction of a full-link intelligent factory, Postny has built a flexible supply chain of "production based on sales"; Nanshan Intelligent Technology has achieved "72-hour rapid response" with 3D design software, and the proportion of custom orders has reached 35%. Behind the flexible manufacturing capability is the deep empowerment of digitalization.
The second is the transformation from single-point breakthrough to systematic integration. Enterprises have moved from the local transformation of "treating the head for a headache and the feet for foot pain" to the systematic reshaping of the entire process, all elements, and all scenarios. Ruo Le Life has deployed the MES/WMS system, and the efficiency of supply chain collaboration has increased by 80%. Data flow, business flow, and value flow are deeply integrated, and technology, organization, and management models are comprehensively reconstructed. The industry is achieving a systematic leap across the entire value chain by building intelligent platforms.
The third is the transformation from closed competition to open collaboration. The digital economy breaks the boundaries of enterprises, industries, and regions, giving birth to brand-new collaboration models. Huafu Technology has launched the "East-to-West Training" model, which connects Shanghai and Urumqi to create a cross-linkage and cross-enterprise computing collaboration value-added space. The Internet of Things, innovative collaboration, and value amplification are becoming the important forms of modern textiles.
3, it is necessary to clarify the application innovation and rich scenarios of artificial intelligence in the textile industry.Artificial intelligence is the biggest variable in the current digital transformation. Since this year, a series of relevant policies have been issued intensively, from the "Opinions on Deepening the Implementation of the 'AI+' Action" to the "Implementation Plan for Further Enhancing the Match between Supply and Demand of Consumer Goods and Further Promoting Consumption". The focus of each place has been promoted, and the intelligent economy has entered the fast lane of development.
The China Textile Federation has identified artificial intelligence as a key area for the future development of the industry, deepening services and guiding practice. Actively promote the popularization and application of AI technology in vertical fields and subdivided scenarios. Carry out special research on "The Path of Digital Transformation and Artificial Intelligence Application in the Textile Industry", release the "Analysis Report on Artificial Intelligence Application in the Textile and Apparel Industry", provide authoritative guidance for the industry; promote tool innovation, optimize the textile product development data analysis and decision-making platform, AI pattern design platform DPI SPACE, empower the development of color, pattern, fabric and terminal products; launch AI application model AiTA, provide high-value data base, knowledge base, model base for the industry; carry out the first AI design competition for cotton clothing in China, the first China (Quxian) outdoor textile and apparel AI design competition and many other competitions, accelerate the popularization and application of technology. The relevant work has achieved positive results. The application of artificial intelligence in the textile industry is rapidly transitioning from concept verification to large-scale implementation, from a single scenario to system integration.
(I) AI-driven design and R&D, creating a digital process for agile innovation in products
AIGC and other new generation AI technologies are changing the traditional design model. Style generation, pattern design, color matching, digital samples and other technologies have significantly increased the speed of design and significantly enhanced the innovation capability of design. 3D digital samples and fabric simulation make sample development faster and cheaper. For example, the large model of Wanshili's pattern design achieves personalized customization, and the style3D platform of Lingdi greatly reduces the sample design cycle, and the Qingdao Light Chain platform greatly reduces the design cost. The trend prediction model of the textile and apparel industry based on big data and AI algorithms helps enterprises to understand market changes. AI is transforming the traditional empirical design model into a new paradigm of "data-driven + intelligent generation".
(II) Integration of AI into intelligent manufacturing to achieve an adaptive precision production model.
Artificial intelligence technology has promoted the production process from experience-driven to data-driven. Predictive maintenance of textile production equipment has effectively reduced downtime and improved production efficiency. The production process is automatically adjusted by AI to optimize the process and achieve energy saving and consumption reduction. The production of apparel enterprises is divided and matched proportionally according to AI algorithms, achieving the minimum allocation of orders and effectively reducing costs, such as the average reduction of 1.5% in raw material costs for apparel enterprises by the intelligent cutting system in Shenzhen Boyi Technology. AI intelligent inspection has solved the problem of defect inspection in weaving enterprises, improving the product pass rate and efficiency, such as the Guangdong Yida edge weaving and inspection system and intelligent fabric inspection machine effectively reducing the rate of defective fabric. AI technology makes the product quality of enterprises more stable, energy consumption lower, and costs more controllable.
(III) AI builds a flexible supply chain to create an efficient and collaborative smart ecology.
AI-driven data enables the textile and apparel supply chain to become flexible, efficient, and collaborative. In the production process, AI vision quality inspection and intelligent scheduling optimize the process flow, enhancing manufacturing capabilities and product quality. For instance, Tongkun Group connects more than 25,000 devices and over 1.5 million key process data collection points, achieving "global visibility" across devices, systems, campuses, and regions. The combination of AI and the "industrial brain" connects information silos, enabling precise fabric matching, flexible capacity scheduling, and full-link data visibility, building an ecological network for deep collaboration between upstream and downstream supply chain partners. This allows for rapid responses guided by real-time market demand, such as the "AI Fabric" large model in Keqiao, which relies on 300,000 fabric data, 40% of the national printing and dyeing industry data, and 28% of the printing and dyeing capacity data to empower the entire industry chain.
(IV) AI empowers brand development and forms a consumer-centered product reach.
AI is deeply restructuring the marketing and consumer reach of textile and apparel brands, accurately predicting fashion demand by analyzing massive data such as social media and search trends, guiding product development and market launch, using AIGC to batch generate personalized marketing content and virtual try-on experiences, achieving creative efficiency and immersive interaction, and building dynamic portraits based on user behavior data, to achieve "thousands of people, thousands of faces" precise reach and recommendations in full channels such as e-commerce and social platforms, building an agile marketing closed loop centered on consumers and intelligent from prediction to reach, significantly improving conversion rates and brand loyalty.
(V) AI promotes green development, and promotes refined management and resource recycling.
AI has become the core driving force for the green development of the textile industry by empowering refined management and technological innovation across the entire chain. In the production process, AI algorithms can optimize energy consumption, precisely control the addition of dyes and auxiliaries, and utilize visual inspection to reduce fabric defects and waste, achieving energy conservation and emission reduction from the source. In terms of resource utilization, AI assists in the development of new circular fiber materials and enhances the efficiency of resource reuse through intelligent monitoring and recycling planning of wastewater and waste materials. From energy conservation optimization to chemical reduction, from carbon footprint management to circular remanufacturing, AI drives the industry to systematically transform from high-energy-consuming traditional models to traceable, recyclable, and sustainable development models.
4, To clarify the development trend of artificial intelligence and the realistic bottlenecks.At present, the rapid innovation of artificial intelligence, quick iteration, and three major evolution trends are emerging, profoundly changing the operation logic and value creation mode of the industry.
(I) Implementation logic "proxy": Business model reconstruction from "traffic entrance" to "service interface". Artificial intelligence is strengthening the ability to understand intentions, decompose tasks, make autonomous decisions, and execute closed loops. Vertical model professional capabilities are continuously improving. Gartner predicts that at least 15% of daily work decisions will be autonomously completed by AI agents by 2028. This will trigger the reconstruction of business logic, such as AI agents replacing humans to make consumption decisions, promoting the transformation of consumption patterns from "search (Search)" to "proxy (Agency)". Enterprises need to explore the transition from focusing on search engine optimization (SEO) to generative engine optimization (GEO) to better establish algorithm trust.
(II) The Extension of the "Embodiment" of Capabilities: From "Understanding the World" to "Transforming the World". The breakthroughs in "Embodied Intelligence" and "Spatial Intelligence" are breaking down the barriers between the virtual and the real. Robots have been successfully applied in actual production processes such as spinning, winding, and cylinder changing. The intelligent patrol robot in the 5G factory of Xin Feng Ming can achieve real-time and accurate equipment status warning. The capabilities of AI are extending from understanding digital spaces to controlling physical entities. Representative spatial intelligence technologies such as Marble system by World Labs and "Puffin" by SenseTime are driving robots from "adapting to programs" to "adapting to environments". In the future, non-standard flexible processes such as sewing and ironing are expected to achieve unmanned operation.
(III) Application Base "Inclusive": From "Machine Replaces Man" to "Human-Machine Coupling" the transformation of production relations. With the iteration of algorithms (such as DeepSeek) and the maturity of the open-source ecosystem, the cost of AI reasoning is declining exponentially, evolving into an "inclusive infrastructure" that is as accessible and payable as electricity. Bain expects that by 2035, the BOM cost of humanoid robot will decrease by 60% to 70%. "Digital employees" are coming true. Human roles are jumping from " execution link" to "supervision and decision-making", and a human-machine relationship of "coexistence, co-governance, and co-prosperity" is being formed. Enterprises need to adhere to people-oriented, improve the governance of human-machine relationship, and release the maximum potential of human-machine collaboration.
But overall, the intelligent economy is still in its infancy and exploratory phase. A MIT report shows that 95% of organizations have not yet obtained substantial value from their AI investment. The development of industry intelligence is still facing bottlenecks. Most enterprises are still in the period of climbing the digital transformation, the digital foundation of management and equipment is not yet solid, and there is a significant " capability generation gap " from the substantial implementation of artificial intelligence. The phenomenon of " data islands " is still prominent, and the key data standards and annotation protocols in the field of industry production are missing, resulting in difficult data collection, difficult large-scale model training, and restricting the improvement of AI model accuracy and the deep development of industry applications. We should not only focus on making up for the shortcomings of digitalization, but also work together to break through the standardization blockage of data circulation and better release the intelligent dividend.
5, To promote the intelligent transformation and original innovation of the textile industry.Looking ahead, the industry needs to establish a "smart native" mindset, adhere to first principles, and systematically promote the digital transformation of the industry.
(I) Promote the coordinated development of industry large models and scenario small models.
The textile industry needs to build a "textile intelligent large model" and form a large number of "small models" in process optimization, defect detection, and scheduling. The large model provides common knowledge and general capabilities, integrates data from the entire industrial chain, and achieves cross-link intelligent decision support. The small model focuses on the fine-tuning of vertical scenarios, and through specific scenarios, it continuously iterates to improve practicality and response accuracy. The two work together to promote the deep implementation and large-scale application of AI in complex and changeable textile scenarios, and to promote the evolution of textile manufacturing towards autonomous perception, dynamic optimization, and predictive maintenance. This provides a solid technical support for the high-quality development of the industry.
(II) Accelerate the evolution from single-point intelligence to full-process collaborative intelligence.
AI will permeate the entire supply chain of the textile and apparel industry, including design, production, logistics, and sales, to achieve cross-link data linkage and intelligent decision-making. Enterprises are no longer limited to single-point improvements. By building an industry-level intelligent system, they can connect upstream and downstream information islands and improve overall operational efficiency. With the improvement of the end-edge-cloud collaborative architecture, real-time and low-latency AI services will be widely used in industry factories, promoting the development of flexible manufacturing and personalized customization models. Changes in consumer demand will accelerate the reverse drive of intelligent restructuring of the production end, and AI will help achieve the paradigm shift from "production-oriented sales" to "demand-oriented production". In the future, the textile industry will gradually form an intelligent ecosystem of "data-driven, model-empowered, and collaborative evolution", continuously releasing productivity and achieving efficient allocation and value reshaping of production factors through technology iteration and application deepening.
(III) Strengthen the deep integration of artificial intelligence and green manufacturing
AI will be deeply integrated into the green transformation of the textile industry, promoting intelligent management of energy consumption, emissions, and resource utilization. By building a carbon footprint tracking model covering the entire industrial chain, the environmental impact assessment of the entire life cycle from raw material procurement to finished product delivery is realized. AI algorithms combined with IoT perception data optimize dyeing and finishing process parameters, reduce water consumption and energy consumption, and reduce waste generation. In the recycling link, intelligent sorting systems improve the reusability of waste textiles, and help closed-loop production. At the same time, a green manufacturing simulation platform is built based on digital twin technology to predict and optimize the ecological benefits of the production line in advance. Policy guidance and technology empowerment work together to promote the textile industry towards low-carbonization and sustainable development, and create a new green productive force.
(IV) Perfecting the open and collaborative AI innovation ecosystem.
The AI empowerment of the future textile and apparel industry requires the upstream and downstream enterprises of the textile industry chain, research institutions, and technology platforms to co-construction and share AI R&D resources, and to promote the open collaboration of elements such as data, algorithms, and computing power. Through new intelligent innovation alliances, accelerate the flow of knowledge and technological innovation, and form an ecological mechanism of collaborative research, result sharing, and risk sharing. The textile industry will accelerate the formation of an innovation model led by leading enterprises, supported by research institutes, and coordinated by small and medium-sized enterprises, to promote the breakthrough of AI technology in key links such as fiber research and development, green printing and dyeing, and intelligent warehousing, and to help traditional manufacturing industry to leap to intelligent, green, and high-end.
