孙怡 腾讯研究院资深研究员 本文为《数字出版研究》2025年第4期刊首语 中华优秀传统文化系民族之根魂，纵观历史，文化创新与文明演进，总与科技的突破相伴而行。如今我 们正见证着千年文脉与数智变革交汇激荡，这场历史性的相逢，将为文明的赓续带来怎样的深刻变革？ 文化传承与创新的数字化、要素化和全球化变革，不仅是技术的考题，更是时代的叩问。 守护珍贵而脆弱的文化遗产，本质上是一场与时间侵蚀的赛跑。当古建面临暴雨险情、甲骨文尚有三千 余字未识、非遗传承人悄然老去，数字技术正以前所未有的方式，有效融入考古、修复与活化的全链 条，为这些濒临失传的记忆构建起跨越时空的"数字基因库"。从甲骨文智能缀合到三星堆文物AI虚拟修 复，再到非遗戏曲的数字新生，大量实践表明，新一代信息技术和前沿交叉技术正推动文化遗产保护突 破既有范式，实现提质增效。科技的意义已远不止于一时抢救，更在于永续的传承和守护。 更重要的是，文化遗产的外溢效应日益凸显，深度融入社会经济大循环。传统文化资源正高效转化为生 产要素，通过沉浸式技术、新型文化装备与创新表达，形成"资源—产品—产业"价值链，催生出文旅消 费的新业态。以游戏《黑神话：悟空》引爆山西古建文 ...

Core Insights - Generative AI (GenAI) is igniting a profound paradigm shift in content production, breaking barriers in high-quality dynamic content generation and pushing complex creative work into the realm of machines [2] - The cultural industry faces both "strategic anxiety" and "opportunity desire" due to the disruptive potential of GenAI, prompting a comprehensive reshaping of existing value chains, business models, and content ecosystems [2] Group 1: GenAI Applications and Industry Transformation - GenAI technology is expected to reduce the production cycle of animated films from three to four years to about one year, and large advertising projects from two to three months to around two weeks, significantly lowering production costs while maintaining or improving quality [9] - The animation industry is transitioning from a labor-intensive model relying on large teams to a new model of lightweight, small teams collaborating with AI, leading to the emergence of new business forms characterized by "AI + high immersion + high sensory experience" [9] - AI-driven animation and short drama markets are anticipated to flourish, with the ability to adapt vast amounts of web literature and comic IPs into diverse styles at unprecedented speeds and lower costs, unlocking significant IP potential [10] Group 2: Structural Evolution of the Animation Ecosystem - A new breed of highly skilled "super individuals" will emerge, possessing top-notch aesthetic and narrative abilities, capable of leveraging AI tools for high-quality creation, replacing traditional large-scale collaborative teams with small, agile groups [11] - Major companies will evolve into "ecosystem builders" providing technology, tools, IP, and channels, while numerous small teams and super individuals will become creative content producers, enhancing overall content supply and quality [11] - The IP industry will see a multidimensional evolution, with GenAI increasing derivative efficiency and market validation speed, while the core standard for enduring IP remains the ability to "occupy user minds" and possess "cross-media narrative capabilities" [12][13] Group 3: Market Dynamics and Content Quality - The market value of real-time generated interactive content varies by application scenario, with gaming being the most promising area due to its non-linear narrative driven by player actions [14] - The acceptance of AI-generated content hinges on quality rather than origin, with the ultimate goal being "technical invisibility," where consumer judgment returns to the content itself [15] - The industry must be vigilant against potential risks posed by GenAI, including over-reliance on AI leading to diminished critical thinking and the risk of creating echo chambers for consumers [16] Group 4: Talent Development and Industry Challenges - Talent cultivation in the industry should focus on foundational skills rather than blind "AI-ification," emphasizing literary, aesthetic, and creative method training to produce individuals who can effectively express ideas using AI [17] - The industry is witnessing a shift towards smaller teams, with a typical configuration of 6-8 members, including specialized roles such as writers, directors, and AI animators, supported by AI technology [25] - The emergence of super individuals and small studios is a mainstream trend, with companies like "With Light and Dust" exploring industrial standards for AI film processes [26] Group 5: Future of IP and Content Creation - The core of IP remains the ability to "occupy minds" and "cross time," with AI facilitating rapid validation of concepts, but the potential for classic IP still relies on deep cultural connections with users [27] - The rise of AI-driven content, particularly in the form of interactive and real-time generated IP, is expected to gain market acceptance as quality improves and becomes indistinguishable from human-created content [29][30] - Companies are actively exploring the integration of AI in content creation, with successful projects demonstrating the commercial viability of AI-assisted original IPs [31]

公众号记得加星标⭐️，第一时间看推送不会错过。 虽然大多数 GenAI 模型都是在大型数据中心集群中进行训练和运行的，但如今在本地构建、测试和 原型化 AI 系统的能力仍然非常重要。 直到最近，这还需要高端的多GPU工作站，价格通常高达数万美元。随着Nvidia在10月份推出基于 GB10核心的DGX Spark，这一现状得以改变。虽然性能远不及后者，但凭借128GB的显存，该系统 本质上是一个内置的AI实验室，几乎可以运行任何AI工作负载。 正如我们在之前的上手体验中所提到的，Spark 并非市面上首屈一指的选择，甚至也不是最便宜的。 AMD 和苹果也提供配备大容量统一内存的系统，这些内存由 CPU 和 GPU 共享，这使得它们在人工 智能开发者和爱好者中广受欢迎。 单，只需按下机器背面的按钮，然后滑开顶盖即可。 然而，由于这两款机器都采用板载 LPDDR5x 内存，因此实际上并没有太多可做的改动。惠普这款机 器配备了两个标准的 2280 PCIe 4.0 x4 M.2 固态硬盘，用户可以自行更换。 相比之下，Spark 更像一台家用电器，不过它的 SSD 也可以通过移除系统底部的磁性板和四个螺丝 来更换。 ...

Core Insights - Generative AI (GenAI) is revolutionizing content production, breaking barriers in high-quality dynamic content generation and pushing complex creative work into the realm of machines [2] - The cultural industry faces both strategic anxiety and opportunity desire due to the disruptive potential of GenAI, prompting a comprehensive reshaping of existing value chains and business models [2] - The "Dawn" research project by Tencent Research Institute and Communication University of China aims to explore the systematic transformation of the cultural industry in the AI era, focusing on applications in long videos, short videos, music, animation, and online literature [2] Group 1: AI Tools and Their Impact - Reading Group has launched AI tools such as Writer Assistant, Comic Assistant, and Copyright Assistant, covering the entire process from writing assistance to IP adaptation [6] - AI cannot replace the emotional and personal expression of excellent writers; the soul and value of a work ultimately depend on human creativity [6][11] - The future online literature ecosystem may present an "olive-shaped" structure, where GenAI serves as a powerful creative "auxiliary wheel," primarily enhancing the "mid-tier" group while the top tier still relies on the talent and effort of writers [6][12] Group 2: Content Creation and Quality - Text and video have structural differences in expression forms, carriers, channels, and audiences, making complete integration unlikely; however, online literature is rapidly evolving into a form that integrates multimodal elements [6][14] - Originality remains the "first principle" of online literature, and the industry must maintain a focus on quality and individual style rather than standardization and maximum efficiency [8][19] - AI tools can assist in visualizing online literature IP and addressing traditional adaptation bottlenecks, but human artistic judgment and decision-making remain central [7][17] Group 3: User Acceptance and Market Dynamics - User acceptance of AI-generated content varies, with some users preferring content created by emotional writers, while others focus on the story itself [20] - The cultural industry must prioritize quality over quantity, as excessive low-quality content can drive users away [19] - The rise of GenAI presents new opportunities for online literature to expand into visual content, enhancing its reach in overseas markets [21][22]

Core Insights - The report titled "Nanjing Primary School Students' Generative AI Literacy White Paper" is the first of its kind in China, focusing on the generative AI literacy of primary school students [1] - It highlights the current state of generative AI application preferences among students and the factors influencing their literacy, revealing phenomena such as "GenAI literacy stratification" [1] Group 1: Survey Findings - Over half of the surveyed students (57.24%) use generative AI as a feedback tool for homework, while 42.28% use it to recommend or generate study materials [2] - A significant portion of students (53.20%) self-learned to use generative AI through online resources, indicating a proactive approach to technology [2] - The frequency of generative AI usage among students is relatively low, with 54.96% using it 1-2 times a week and 17.31% using it daily [2] Group 2: Limitations in Application - Despite familiarity with generative AI, students primarily use it for basic tasks like information retrieval and homework correction, lacking deeper integration with complex problem-solving [3] - More than 80% of students exhibit a strong critical awareness of the content generated by AI, yet nearly half struggle with identifying inaccuracies due to a lack of effective strategies [3] - Students face challenges in content verification and critical analysis, particularly with complex or obscure generated content [3] Group 3: Ethical Concerns - Some students choose to conceal their use of generative AI due to concerns about punishment or ethical boundaries, highlighting a disconnect between school evaluation systems and the widespread use of AI [4] Group 4: Recommendations for Improvement - The white paper suggests integrating generative AI literacy into the broader AI education framework in schools, including dedicated courses on AI principles and critical thinking [5] - Schools are encouraged to establish clear guidelines for the academic use of generative AI, helping students navigate ethical dilemmas [5] - Collaboration with professional institutions for workshops and public awareness activities is recommended to promote proper generative AI application [5] Group 5: Future Implications - The white paper is expected to advance the development of AI education in Chinese primary and secondary schools towards a more scientific, equitable, and high-quality approach [6]

Core Insights - The report highlights the current state of elementary school students' engagement with Generative AI (GenAI) in Nanjing, indicating a significant level of interaction and awareness among students regarding its use in learning [1][2]. Group 1: Current Usage and Awareness - 45.35% of surveyed elementary students discuss learning challenges with GenAI, while 62.06% are aware they need to inform teachers about their AI usage, but only 32.35% do so consistently [1]. - 57.24% of students use GenAI for feedback on assignments, and 42.28% utilize it for generating learning materials [2]. - 53.20% of students self-learned to use GenAI through online resources, with 54.96% using it 1-2 times a week [2]. Group 2: Limitations in Deep Learning - Many students primarily use GenAI for basic tasks like homework checking and information retrieval, lacking deeper integration with complex problem-solving [3]. - Over 80% of students show a strong critical awareness of GenAI-generated content, yet nearly half struggle with identifying inaccuracies due to a lack of effective strategies [3]. Group 3: Ethical Concerns and Educational Gaps - Some students conceal their use of GenAI due to fears of punishment or ethical ambiguity, indicating a disconnect between school evaluation systems and the widespread use of GenAI [4]. - The report suggests integrating GenAI literacy into the AI education framework in schools, emphasizing the need for clear guidelines on AI usage in assignments [5]. Group 4: Recommendations for Improvement - The report recommends establishing a collaborative educational system involving schools, parents, and professional institutions to enhance GenAI literacy [6]. - It advocates for the development of GenAI literacy courses in schools to improve students' skills in critical thinking, academic integrity, and ethical decision-making [5][6].

Core Viewpoint - Generative AI (GenAI) is igniting a profound paradigm shift in content production, breaking down barriers to high-quality dynamic content generation and pushing complex creative work into the realm of machines. This technological advancement brings both strategic anxiety and opportunity to the cultural industry, prompting a comprehensive rethinking of existing value chains, business models, and content ecosystems [2]. Group 1: Application of GenAI - In fields like online literature and music, GenAI is widely applied throughout the entire production process, with platforms embedding easily accessible AI generation tools, leading to generalized and socialized creative capabilities. The industry widely believes that content creation should adhere to "human-machine collaboration" while enhancing production efficiency through "engineering" [7]. - GenAI's fundamental difference from previous technologies lies in its potential to replace certain human capabilities, evolving into a "new species" that competes directly with humans. AI-generated content will "eliminate mediocrity," forcing human creators to strive for higher quality, shifting the industry from "quantity competition" to "quality competition" [7]. - The emergence of "super individuals" or "micro-teams" will become the new norm, with "human-machine collaboration" as the core competitive advantage. Future content producers must be adept at harnessing AI, acting as "directors" or "architects" in the creative process [7]. Group 2: Impact on Cultural Industry - GenAI will disrupt the existing interests within the cultural industry, with copyright confirmation and revenue distribution becoming core challenges and significant opportunities for reshaping the industry. The potential for "super individuals" to bypass intermediaries and connect directly with consumers may lead to new business models [8]. - Consumer acceptance of AI-generated content hinges on content quality. GenAI is driving a shift in consumer motivation from superficial "emotional stimulation" to deeper "emotional and value recognition," creating a new blue ocean of content composed of numerous small yet exquisite IPs [8]. - The traditional "talent growth path" in the content industry may face disruption due to GenAI, which excels in "diversity" but poses challenges in "controllability." There is a need to be cautious about AI eroding the significance of creation and the soil for talent growth [9]. Group 3: Insights from Industry Experts - Industry experts emphasize that while GenAI is making strides in various cultural content forms, the actual implementation of "cost reduction and efficiency enhancement" in content production remains to be fully realized. The current capabilities of GenAI are still limited, and human creators will continue to play a crucial role in high-quality outputs [10]. - The music industry is witnessing a significant shift, with many companies adopting AI for music creation and production processes. However, while AI can generate music, it still relies heavily on user input and creativity to achieve desired results [11]. - The concept of "content engineering" is gaining traction, where the creative process is standardized and can be automated to a degree, allowing for rapid production of content while still requiring human creativity for high-quality outcomes [12]. Group 4: Future of Content Production - The future landscape of content production may see a shift towards direct engagement between creators and platforms, with the potential for individual creators to establish their own brands and sell their works directly to consumers [24]. - The emergence of new roles in the music industry, such as those who can effectively collaborate with AI tools, will be crucial. The industry may see a rise in "bedroom musicians" who can independently create and monetize their music using AI [20]. - The acceptance of AI-generated content by consumers will depend on the perceived quality of the output. As AI-generated works improve, consumers may become indifferent to whether content is created by humans or machines, leading to a potential oversaturation of average-quality content [27][28]. Group 5: Concerns and Challenges - There are concerns that the rise of AI in content creation may lead to a lack of growth opportunities for emerging creators, as reliance on AI could hinder the traditional learning and development processes necessary for becoming skilled authors [31]. - The music industry may face significant challenges as AI-generated music becomes more prevalent, potentially displacing many current musicians and altering the landscape of music creation [32]. - The relationship between human creativity and machine-generated content presents a "impossible triangle" scenario, where achieving low labor costs, low machine costs, and high-quality output simultaneously may not be feasible [33].

Core Insights - The global agriculture sector is at a critical juncture, facing unprecedented pressures from climate change, resource degradation, demographic shifts, and geopolitical instability, necessitating a systemic transformation led by "deep technology" [1] - Deep technology, which encompasses advanced scientific and engineering innovations, is expected to revolutionize the agricultural industry and address significant global challenges over the next decade [1] Group 1: Deep Technology in Agriculture - Deep technologies such as Generative AI, computer vision, edge IoT, satellite remote sensing, robotics, CRISPR gene editing, and nanotechnology are identified as key drivers for transforming global agriculture into a more resilient, sustainable, and efficient system [1] - The World Economic Forum's "AI in Agriculture Innovation Initiative" released a report highlighting the potential of these technologies to reshape agricultural practices [1] Group 2: Generative AI - Generative AI is leveraging advancements in large language models and the increasing availability of agricultural data, providing personalized crop management advice and localized farming plans [2] - Applications include acting as an "AI advisor" for farmers, assisting governments in macro crop planning, and accelerating the development of new crop varieties through gene editing [2] - The lack of high-quality training data, particularly for localized scenarios, remains a significant barrier to the widespread adoption of Generative AI in agriculture [2] Group 3: Computer Vision - Computer vision enables machines to interpret images and videos, generating decision-making suggestions and reducing reliance on human analysis [3] - In agriculture, it is used for precise identification of crop diseases, weeds, and pests, as well as real-time monitoring of crop growth [3] - The variability of field conditions and plant growth stages poses challenges for the large-scale application of computer vision technology in agriculture [3] Group 4: Edge IoT - Edge IoT processes data at the device level or nearby network edge, allowing for low-latency real-time responses and accelerating autonomous decision-making [4] - It is particularly beneficial in rural areas with weak network coverage, facilitating applications such as automated irrigation and early disease warning systems [4] - High equipment costs and interoperability issues between different edge systems are current challenges in this field [4] Group 5: Satellite Remote Sensing - Satellite remote sensing technology is increasingly applied in agriculture due to improved spatial and spectral resolution and higher data collection frequency [6] - It allows for efficient monitoring of large geographic areas at a low cost, assessing crop health and predicting pest outbreaks [6] - The precision of satellite remote sensing needs improvement when dealing with small-scale, dispersed farmland or multi-crop rotations [7] Group 6: Robotics - Robotics technology automates labor-intensive or complex tasks in agriculture, integrating perception and decision-making capabilities [8] - With advancements in AI perception and cloud-edge collaboration, agricultural robots can perform tasks such as precision planting and automated harvesting [8] - High costs of these technologies present challenges for adoption in countries with abundant low-wage labor [9] Group 7: CRISPR Technology - CRISPR gene editing is a key force in agricultural development, allowing precise modifications to DNA to enhance desirable traits in crops [10] - It aims to accelerate the breeding of crops that are drought-resistant, pest-resistant, and nutritionally enhanced [10] - Regulatory hurdles and public acceptance issues are significant challenges to the commercialization of CRISPR technology [11] Group 8: Nanotechnology - Nanotechnology shows potential in agriculture for pest control, nutrient management, and controlled release of agricultural inputs [12] - The lack of long-term data on environmental and health impacts poses challenges for the widespread application of nanotechnology [12] - The report suggests that governments and institutions should support promising agricultural deep tech projects through policy coordination, funding, talent development, and infrastructure building [12]

Core Insights - Generative AI (GenAI) is igniting a profound paradigm shift in content production, breaking down barriers to high-quality dynamic content generation and pushing complex creative work into the realm of machines. This technological advancement brings both "strategic anxiety" and "opportunity desire" to the cultural industry, prompting a reevaluation of existing value chains, business models, and content ecosystems [2] Group 1: GenAI's Impact on Content Production - GenAI has penetrated various cultural content production processes, with varying degrees of involvement across different segments. It can effectively replace repetitive labor and high-cost production stages, but it cannot achieve cost reduction and efficiency in all areas, as some tasks still outperform machines [6] - The overall scale of AI-native content is expected to grow rapidly, particularly in areas like AI short videos and AI comics. As GenAI's capabilities expand, new workflows of "human-machine collaboration" will emerge, leading to real-time dynamic content generation that meets consumer demands instantaneously [6][12] - GenAI empowers individual content creators, leading to the emergence of new types of producers characterized by individualization, small scale, and cross-domain collaboration. While social specialization will change due to AI, the concept of "division of labor" will persist, with specialized content producers mastering "human-machine collaboration" becoming mainstream [6] Group 2: Changes in IP and Business Models - The traditional IP operation models, copyright definitions, and profit distribution mechanisms in the cultural industry will undergo changes, with specific attempts already observed in the short video sector. However, comprehensive industry transformation will require further exploration [6] - The concept of copyright may fundamentally change, with potential new models emerging where content is not owned by a single entity but rather shared among participants. This necessitates new rules and legal frameworks [20] - The commercial ecosystem driven by AI will undergo a fundamental restructuring, shifting from explicit advertising to on-demand production based on user desires. This could lead to the emergence of transient IPs that exist only for short periods to meet immediate sales goals [20] Group 3: Consumer Acceptance and Concerns - Consumers are likely to accept AI-generated content as long as it meets their basic quality standards. New payment models may arise based on whether content satisfies individual consumer needs, with GenAI potentially raising the average quality of content and eliminating inferior offerings [7][21] - Concerns exist regarding the ability of GenAI to replace the traditional learning and training processes required for developing professional talent in the industry. The controllability of GenAI's capabilities is also a significant concern [7][25] - The current challenge lies in the insufficient capabilities of generative AI, which may lead to a harsh price war in the industry, resulting in an influx of homogeneous, low-quality content that could overshadow high-quality productions [25]

Core Insights - The 7th China Supercomputing Conference concluded with Lenovo achieving the top position in the HPC TOP100 list for the 11th consecutive time, showcasing its strong capabilities in computing infrastructure [1][2] - Lenovo was recognized as a "Leading Enterprise in Computing Power" and its innovative solution was included in the "2025 China Computing Power Application Classic Cases" [1] - The conference introduced the concept of "computing power economy," emphasizing the direction and pathways for the new computing power industry [2] Computing Power Demand - There is a significant shift in market demand from training to inference and post-training phases, leading to explosive growth in inference computing power requirements [3] - Gartner predicts that global spending on generative AI will reach $644 billion by 2025, a 76.4% increase from 2024, with approximately 80% allocated to AI hardware for inference scenarios [3] Lenovo's Technological Innovations - Lenovo has made substantial advancements in server hardware innovation and AI computing optimization, ranking among the top three in China's AI server market sales in the first half of 2025 [4] - The company has introduced a comprehensive range of servers, including data processing, AI training, and inference application servers, to meet the evolving demands of AI applications [4] Industry Applications - Lenovo is driving the large-scale implementation of AI across various industries, including manufacturing, education, and finance, with notable benchmark cases [5] - The "Blue Whale No. 1" liquid-cooled high-performance computing platform was established for Nanjing University, demonstrating high space utilization efficiency with 360 computing nodes and two high-performance storage systems [5] Future Outlook - The year 2025 is anticipated to be crucial for the evolution of global AI infrastructure, with Lenovo's China Infrastructure Business Group planning to expand its business scenarios [6] - The company aims to create a robust computing service engine focused on "AI-driven" and "localized" strategies to support the intelligent transformation of various industries [6]