2026年行业展望 1、模型平权趋势加速，国产算力渗透率持续提升。当前国产大模型密集走开源路线，虽然相比国际闭 源模型仍有差距，但凭借性价比优势市场份额快速增长。根据SuperCLUE中文大模型基准测评9月月 报，排名前20的大模型中，国产开源大模型有9个，接近一半比例。国内模型平均API价格为3.88元/百 万tokens，海外模型平均API价格为20.46元/百万tokens，海外模型的价格远高于国内模型。伴随模型平 权，推理算力需求强劲增长叠加美国对高端芯片封锁，国产AI芯片需求保持高速增长，算力供需剪刀 差仍在扩大。 2025年回顾，计算机行业呈现"先扬后抑、震荡轮动"的走势，年初至今跑赢沪深300 2025年计算机行业指数整体呈现"先扬后抑、震荡轮动"的走势，截至11月18日，行业指数年内累计上涨 19.57%，小幅跑赢沪深300指数。财务层面，2025年前三季度行业呈现"营收小幅回暖、利润高增，现 金流仍紧"特征，行业整体前三季度实现营收同比增长9.13%，归母净利润同比增长38.25%，毛利率同 比下降1.19%，净利率同比上升0.50%，应收账款周转率同比增长8.17%;当前计算机行业指数估值 ...

Group 1 - The core viewpoint highlights that the Hong Kong technology sector is undergoing a critical phase of value reassessment, driven by a surge in domestic generative AI user adoption, with over 90% of users preferring local models [1] - Major Hong Kong tech companies are not only users of AI technology but also key builders of the industry chain, covering high-growth areas such as software and hardware, new energy vehicles, and innovative pharmaceuticals [1] - The CSI Hong Kong Stock Connect Technology Index shows a significantly lower price-to-earnings ratio compared to the A-share ChiNext Index, with a horizontal discount exceeding 40%, while the vertical AH share premium index continues to converge [1] Group 2 - In terms of financial performance, Midea Group reported a net profit attributable to shareholders of approximately 37.883 billion yuan for the first three quarters, reflecting a year-on-year growth of 19.51% [1] - The influx of southbound funds and global capital is creating a resonance that drives the valuation recovery of the technology sector [1] - The technology sector in Hong Kong is seen as an ideal choice for investing in "new quality productivity" due to its growth potential and valuation advantages in the fourth quarter [1]

Core Insights - The transition from the "digital age" to the "intelligent age" is underway, with significant growth in the user base of generative artificial intelligence (AIGC) in China, reaching 515 million users by June 2025, a penetration rate of 36.5% [1] - The rapid increase in users, up by 266 million or 106.6% compared to the end of last year, indicates that the market is on the brink of an explosion [1] Group 1: User Growth and Technology Maturity - The explosive growth in user numbers is attributed to technological maturity and a rich array of products, with over 90% of users preferring domestic large models [1] - A total of 538 generative AI services have been registered in China by August 2025, with applications spanning various fields such as Q&A, office tasks, entertainment, and content creation [1] Group 2: Systemic Restructuring of Industries - AIGC is driving a systemic restructuring of production logic from "process-driven" to "human-machine collaboration," allowing AI to assist in knowledge and creative tasks, thereby redefining innovation speed and cost in industries like pharmaceuticals and engineering [2] - The industrial organization is evolving from a "chain-based ecosystem" to a "networked ecosystem," enabling decentralized production and content creation, lowering barriers for small teams and individuals to produce professional-quality outputs [3] - Competitive logic is shifting from "scale" to "ecosystem," where companies that can create a closed loop of user engagement and data optimization will establish a dynamic competitive advantage [3] Group 3: Challenges and Ethical Considerations - The rise of AIGC also brings forth challenges such as the emergence of a new "digital divide," where disparities between those who effectively utilize AI and those who do not become pronounced [4] - Ethical and regulatory challenges are intensifying, with concerns over AI bias, data privacy, and accountability becoming pressing social issues as the user base expands [4]

Economic Performance - In the first three quarters, China's GDP reached 101.5 trillion yuan, with a year-on-year growth of 5.2% [1] - This growth rate positions China among the top economies globally, with the economic total in Q3 reaching 35.5 trillion yuan, surpassing the projected total for the third-largest economy in 2024 [1] Industrial Growth - Significant growth in new industries and technologies is evident, with industrial robots, service robots, and high-speed trains seeing production increases of 29.8%, 16.3%, and 8.6% respectively [2] - The manufacturing value added in lithium-ion battery production, shipbuilding, and motor manufacturing grew by 29.8%, 22.9%, and 17.1% respectively, indicating a robust transformation towards new production capabilities [2] Consumer Spending - Final consumption expenditure contributed 53.5% to economic growth, an increase of 9.0 percentage points compared to the previous year [2] - The implementation of a long-term special government bond policy supporting the replacement of old consumer goods has shown effectiveness, with over 8.3 million applications for vehicle replacements by September 10 [2] Green Transition - The production of new energy products such as electric vehicles and solar batteries has seen substantial growth, with increases of 29.7%, 46.9%, and 14.0% respectively [3] - The growth in wind and solar power generation also reflects a strong commitment to green transformation, with both sectors achieving double-digit growth [3] Economic Indicators - Positive indicators are accumulating, with the manufacturing purchasing managers' index rising for two consecutive months as of September [3] - The domestic travel during the Mid-Autumn and National Day holidays reached 888 million trips, indicating strong consumer engagement and activity [3]

Core Insights - The report indicates that by June 2025, the user base for generative artificial intelligence in China is expected to reach 515 million, with a penetration rate of 36.5% [1] - The next phase of development for generative artificial intelligence is anticipated to yield new results in five areas: model integration, open-source community contributions, embodied intelligence for new user interactions, expansion of AI capabilities, and improved governance [1] User Growth and Demographics - The user base for generative artificial intelligence doubled in the first half of the year, increasing by 266 million users, representing a growth rate of 106.6% [2] - The core user demographic consists of young individuals under 40, who make up 74.6% of users, and those with higher education (college degree or above) account for 37.5% [2] Domestic Model Preference - Over 90% of users prefer domestic generative AI models, reflecting the maturity and improved user experience of these products [2] - Generative AI is facilitating quality enhancement and efficiency improvements across various sectors [2] Regulatory and Application Landscape - As of August 2025, a total of 538 generative AI services have been registered, with 263 applications or features officially documented [3] - Generative AI is being widely applied in areas such as intelligent search, content creation, office assistance, and smart hardware, with ongoing exploration in agriculture, industrial manufacturing, and scientific research [3] Industry Development and Trends - The AI industry in China is expanding, with a complete industrial system covering foundational, framework, model, and application layers [4] - China leads globally in AI patent applications, with 1.576 million applications accounting for 38.58% of the global total as of April 2025 [4] - Embodied intelligence has emerged as a development hotspot, with significant investment and application in education, healthcare, and offline services [4]

Core Insights - The report by the China Internet Network Information Center indicates that as of June this year, the user base for generative artificial intelligence in China reached 515 million, with a penetration rate of 36.5% [1] - Domestic enterprises are actively investing in the research and development of generative AI technologies, with over 90% of users preferring domestic large models [1] - The application scenarios for generative AI are continuously expanding, with significant contributions to various fields such as agriculture, industrial manufacturing, and scientific research [1] Industry Development - China is strengthening its foundational research in artificial intelligence while emphasizing application-oriented approaches, leading to deeper penetration of generative AI technologies into specific application scenarios [2] - A complete artificial intelligence industrial system has been established in China, covering all relevant aspects including chips, computing power, data, platforms, and applications [2] - As of April this year, China has filed 1.576 million AI patent applications, accounting for 38.58% of the global total, positioning the country as a leader in AI patent filings [2]

Macro Perspective - The current bull market is supported by a macro environment similar to that of 2014-15, driven by technological innovation, policy support, and ample liquidity[10] - The bull market is not primarily driven by significant improvements in economic fundamentals, as GDP growth remains low and corporate profit growth is also weak[10] - Current market indicators show that while valuations are high, there is still upward space compared to the peak values of the previous bull market[2] Industry Insights - The technology sector, particularly AI, is expected to lead market trends, with significant opportunities in TMT (Technology, Media, and Telecommunications) and advanced manufacturing[3] - The electronics industry has seen a cumulative increase of 121.0% since September 2024, driven by AI innovation and demand for computing power[39] - The computer sector is benefiting from policy and technological improvements, with a focus on AI-related investment opportunities[3] Investment Recommendations - Key stocks in the semiconductor sector include SMIC and Hua Hong Semiconductor, while the consumer electronics sector highlights companies like Luxshare Precision and Zhuhai Conpuc[42] - In the AI computing space, recommended stocks include Haiguang Information and Cambrian[42] - The gaming sector within media is also promising, with leading companies like 37 Interactive Entertainment and Gigabyte Networks showing strong market share growth[3] Market Outlook - The mid-term outlook for the equity market remains optimistic, with a focus on three main growth lines: AI, advanced manufacturing, and new consumption driven by domestic demand[3] - The current bull market is expected to continue, with the potential for further capital inflows into the technology sector as the global AI industry evolves[3]

Core Viewpoint - The Chinese government has issued a strategic document to promote the integration of artificial intelligence (AI) into various industries, marking a significant step towards an intelligent economy and society. Group 1: AI Industry Development - AI is accelerating the transition of the national economy towards intelligence across three major industries [1] - The AI industry is undergoing a global restructuring, with a shift from isolated technological breakthroughs to systematic innovation [1] - The emergence of domestic AI models has positioned China among the top tier internationally, creating a scalable commercial application era [1] Group 2: Implementation Pathways - The document outlines a three-stage goal for AI integration: widespread application in key areas by 2027, formation of significant growth poles by 2030, and full transition to an intelligent economy by 2035 [2] - The strategy emphasizes sector-specific policies, focusing on six areas: scientific technology, industrial development, quality improvement in consumption, public welfare, governance capabilities, and global cooperation [2] Group 3: Core Objectives and Future Outlook - The core objective is to deeply integrate AI with the real economy across all elements, chains, and scenarios, transforming traditional industries and creating new "AI-native" business models [3] - The "AI-native" concept refers to businesses built entirely on AI infrastructure and logic, which are expected to dominate the future economy [3] - The initiative is seen as a comprehensive and strategic guide to accelerate AI application across various sectors, driving quality, efficiency, and dynamic changes in industries such as manufacturing, services, finance, and transportation [3]

Group 1 - The article emphasizes five key sectors with marginal improvements to focus on in August: AI applications, AI hardware, non-bank financials, defense and military industry, and innovative pharmaceuticals [1] - From a medium to long-term perspective, the strategy suggests focusing on cycles and supply-demand dynamics, particularly in the context of the new technology cycle and the progress of societal intelligence [1] - Key areas of attention include the continuous iteration of large models, the enhancement of computing infrastructure and AI ecosystem, the realization of AI business models, and the empowerment of consumer electronics and robotics by AI [1] Group 2 - The article highlights the importance of domestic substitution cycles, particularly in relation to self-controllable industrial chains such as domestic large models, AI applications, computing power, and integrated circuit industry chains [1] - It also discusses the "dual carbon" cycle, focusing on cost reduction and efficiency improvement across the entire carbon neutrality industrial chain, including solar power, wind energy, energy storage, hydrogen energy, and nuclear power [1] - The trend towards electrification and intelligence is noted, with an increasing penetration rate of electric smart vehicles [1]

Group 1 - The core value of Guangzhou's digital economy is projected to grow steadily in 2024, with the core industry value exceeding 400 billion yuan for the first time, reaching 447.03 billion yuan, a year-on-year increase of 9.8%, accounting for 14.41% of the city's GDP [1] - The contribution rate of the digital economy to the city's economic growth is 61.3%, indicating its significant role in optimizing industrial structure and fostering new economic momentum [1] - The rapid evolution of artificial intelligence technology is reshaping reality and defining the future, becoming a strategic technology leading a new round of technological revolution and industrial transformation [1] Group 2 - The development of artificial intelligence has transitioned from symbolic systems to large model eras, characterized by multi-modal, multi-scenario, and multi-task capabilities, driven by computing power and data [2] - Domestic large models are widely applied in various fields such as public services, biological sciences, finance, smart buildings, human resource management, and psychological therapy, although challenges like AI hallucinations and privacy protection need to be addressed [2] - The humanoid robot sector is currently facing challenges such as high costs, weak regulation, and limited application scenarios, with future development focusing on low-cost, high-efficiency remote-controlled robots for commercial data collection [2]