国泰海通晨报 2025 年 10 月 09 日 国泰海通证券股份有限公司 研究所 [Table_Summary] 1、【海外策略研究】：①年初以来港股表现出色，虽三季度港股阶段性跑输 A 股，但 9 月以来在 科技股驱动下已重拾升势。②当前港股估值仅处历史中位，与 A 股对比性价比仍然突出，其中科 技板块低估优势更加明显。③产业趋势叠加资金面持续改善，港股四季度有望续创新高，结构上 港股科技仍是行情主线。 2、【固定收益研究】：国债发行计划变化、制造业 PMI 成色和海外资产涨跌。 [汤蔚翔 Table_Authors] (分析师) 电话：021-38676172 邮箱：tangweixiang@gtht.com 登记编号：S0880511010007 [Table_ImportantInfo] 今日重点推荐 吴信坤(分析师) 021-23154147 wuxinkun@gtht.com S0880525040061 杨锦(分析师) 021-23185661 yangjin2@gtht.com S0880525040083 陆嘉瑞(研究助理) 021-23185659 lujiarui@gtht.com S0 ...

Core Insights - The AI sector is experiencing significant growth, with the Shanghai Stock Exchange's Sci-Tech Innovation Board AI Index rising by 1.28% as of October 9, 2025, driven by strong performances from key stocks like Chipone Technology and Lattice Semiconductor [2] - OpenAI has made notable advancements, including the launch of its video generation model Sora 2.0, which is expected to revolutionize the video generation field [4] - Citigroup analysts have raised their AI capital expenditure forecast for 2026 from $420 billion to $490 billion, indicating a growth expectation of 24%, surpassing the current market consensus of 20% [3] Industry Developments - Ant Group's AI model Ring-1T, with a parameter count of 1 trillion, was announced as open-source on September 30, 2025, highlighting advancements in AI model capabilities [2] - OpenAI's partnership with major chip manufacturers like Samsung and SK Hynix for the "Stargate" project signifies a strategic move to enhance hardware support for AI applications [4] - The AI video generation market is projected to reach a trillion-dollar valuation, driven by the successful integration of social features in Sora 2, which has quickly topped the U.S. iOS app charts [6] Market Performance - The Sci-Tech AI ETF (588790) has shown strong performance, with a 2.56% increase over the last two weeks and a daily average trading volume of 747 million yuan over the past month, ranking first among comparable funds [2][8] - The ETF's assets have grown by 2.642 billion yuan over the past three months, indicating robust investor interest and confidence in the AI sector [8] - The top ten weighted stocks in the Sci-Tech AI Index account for 71.9% of the index, with companies like Lattice Semiconductor and Chipone Technology leading the way [9]

Market Overview - The market remained relatively calm before the National Day holiday, with the Shanghai Composite Index and Shenzhen Component Index closing up 0.52% and 0.35% respectively, while the ChiNext Index remained flat [4] - Various sectors showed strong performances, including AI video, storage, and non-ferrous metals, with several stocks hitting the daily limit up [3] AI Video Sector - OpenAI plans to release a new version of the Sora video generator as an independent application, leading to significant stock price increases for companies like Wangda Software and Danghong Technology, both hitting the daily limit up [11] - The Sora AI video sector saw a resurgence after a long period of inactivity, with related stocks experiencing notable gains [11] New Energy Vehicles - The new energy vehicle sector showed strong performance, with multiple stocks such as Songyuan Safety and Fulin Precision hitting the daily limit up [12] - The sector's growth is largely driven by lithium battery concepts, particularly solid-state batteries, with companies like Seres and Dongfeng Motors seeing substantial gains [12] Atomic-Level Manufacturing - The Ministry of Industry and Information Technology is considering including atomic-level manufacturing in the upcoming 14th Five-Year Plan, which is expected to create investment opportunities in this strategic sector [7] - Stocks like Guolin Technology and Del Future saw significant price increases following the announcement, reflecting market interest in the new direction outlined in the plan [7] Technology Sector - Fangzheng Technology received a large order from NVIDIA, leading to a significant stock price increase, with the company expected to see substantial performance improvements in the second half of the year [15] - The stock's rise was also influenced by market rumors regarding competitors, indicating a competitive landscape in the technology sector [15][16]

Investment Rating - The report maintains a "Recommendation" rating for the computer industry, expecting the industry index to rise more than 5% over the next 3-6 months compared to the benchmark index [15]. Core Insights - The release of DeepSeek V3.2-Exp model is expected to drive the development of the domestic AI ecosystem, showcasing the synergy between domestic chips and large models, which leads to significant improvements in computational efficiency and cost reduction [5]. - The introduction of the DeepSeek Sparse Attention mechanism allows for a substantial enhancement in long text processing efficiency, overcoming traditional computational complexity limitations [5]. - The cost of AI applications is anticipated to decrease significantly, with the DeepSeek API prices reduced by over 50%, making advanced AI models more accessible to developers and businesses [5]. - Investment focus is recommended on domestic chip manufacturers like Cambricon and Hygon, as well as companies within the Huawei supply chain and various AI application firms [5]. Industry Basic Data - The computer industry comprises 337 listed companies with a total market capitalization of approximately 61,619.92 billion and a circulating market capitalization of about 55,710.57 billion [2]. Relative Index Performance - The absolute performance of the computer industry over the past 1 month, 6 months, and 12 months is -1.6%, 24.3%, and 72.1% respectively, while the relative performance is -4.3%, 6.3%, and 47.3% [3].

Group 1: Market Performance - The non-ferrous metals, storage chips, and AI application sectors led the market gains today, with significant increases in key stocks such as Luoyang Molybdenum, Huayou Cobalt, Jiangxi Copper, and Northern Rare Earth [1][2] - The Shanghai Composite Index rose by 0.4%, the Shenzhen Component Index increased by 0.31%, and the ChiNext Index saw a slight rise of 0.06% [1] Group 2: Non-Ferrous Metals Sector - The non-ferrous metals sector experienced a strong rally, driven by energy metals, industrial metals, and minor metals [2] - Key catalysts for this sector include a policy plan from the Ministry of Industry and Information Technology aiming for an average annual growth of 5% in the non-ferrous metals industry from 2025 to 2026, and a projected annual growth of 1.5% in the production of ten non-ferrous metals [4] - The Federal Reserve's recent interest rate cut has led to expectations of a new round of easing, further supporting the sector [4] - Supply-side disruptions have also contributed to the sector's performance, with a focus on copper and aluminum as key investment opportunities [5] Group 3: AI Applications Sector - The AI applications sector showed strong performance, with stocks like Danghong Technology and Yidian Tianxia experiencing significant gains [7] - The Sora concept and AI corpus sectors also saw increases, reflecting growing interest and investment in AI technologies [8] - Recent developments in AI models, such as the release of DeepSeek-V3.2-Exp and Anthropic's Claude Sonnet 4.5, indicate ongoing advancements in the field [9][10]

Core Insights - The article discusses the collaboration between ArcSoft and Qualcomm at the Snapdragon Summit, highlighting the launch of the Snapdragon 8 Elite Gen5 platform and the introduction of ArcSoft's Video Dragon Fusion technology aimed at enhancing video quality for mobile devices [1][2]. Group 1: Technology and Innovation - ArcSoft's Video Dragon Fusion is a comprehensive solution that leverages hardware collaboration and algorithm innovation to overcome industry challenges in achieving cinematic video quality [2]. - The technology demonstrates superior performance in dynamic range, tonal transitions, and color reproduction compared to mainstream solutions [2]. - The collaboration has led to a 40% overall performance improvement and a 20% reduction in power consumption for the technology, thanks to deep optimization for Qualcomm's ISP and AI engine [3]. Group 2: AI and User Experience - The role of mobile AI is evolving from merely optimizing image quality to acting as a creative assistant, as showcased by ArcSoft's various AI imaging products at the Snapdragon Summit [4]. - The AgenticAI video creation feature allows users to generate stylized short videos by simply inputting natural language commands, making video production accessible to ordinary users [4]. - ArcSoft's 3D photo generation technology transforms static 2D portraits into dynamic 3D representations, ensuring user privacy through local data processing while creating personalized 3D image assets [4]. Group 3: Future Directions - ArcSoft aims to continue exploring AI applications in various imaging scenarios beyond smartphones, including smart cars, AI glasses, robots, and smart cameras, to enhance image quality and lower technology usage barriers [4].

Investment Rating - The report suggests a focus on leading domestic generative large model companies such as iFlytek, AI hardware companies like Hikvision, Hongsoft Technology, and Hesai, as well as companies like Maifushi that can enhance paid rates and ARPU values [2] Core Insights - The report highlights significant advancements in AI models, with Alibaba launching seven new models at the 2025 Yunqi Conference, enhancing performance and efficiency [4][11] - The overall operating strength in the second half of the year is expected to improve compared to the first half, driven by low baselines and accelerated technology implementation [11] - The report identifies high certainty investment directions in AI, including emerging hardware for edge AI, software for C-end markets, enterprise services, and large model deployments for G-end clients [11] Summary by Sections 1.1 Computer Industry Insights - The AI industry chain is experiencing high demand in computing power and software, with significant year-over-year growth in AI model bidding amounts [12] - The laser radar sector maintains high demand, with a notable increase in installation volumes for ADAS [12] 1.2 Subsector Insights - High demand sectors include AI computing power and laser radar, while sectors like industrial software and medical IT are facing some pressure [12] - The report anticipates steady growth in software outsourcing, quantum computing, and data elements, with a positive outlook for education IT and security sectors [12] Market Performance Review - From September 22 to September 26, 2025, the computer industry index rose by 0.61%, outperforming the CSI 300 index by 0.62 percentage points [14] - The top five performing companies in the computer sector during this period included Pinming Technology and Zhongwei Electronics [15] Upcoming Key Events - Significant upcoming events include the 2025 Industrial AI Co-Promotion Conference and OpenAI's Developer Conference, which may present investment opportunities in related sectors [26]

新华社杭州9月27日电特稿｜当下，AI如何赋能跨境电商？ 新华社记者郭文迪孙晶魏一骏 当商家想发布信息时，只需上传一张图片，商品的标题、详情、属性就会自动生成；在营销投放环 节，各类广告素材也能自动生成；在售后服务环节，商家和全球消费者的交流7×24小时不间断……在 人工智能（AI）帮助下，这样的跨境电商运营场景当下已不难实现。 25日，在杭州开幕的第四届全球数字贸易博览会上，许多跨境电商领域展商着力推广自身引以为傲 的AI技术，积极讲述AI在经营中带来的精彩故事，呈现出当下AI技术赋能跨境电商的火热场景。 机遇：向"AI+跨境"模式跃迁 据阿里巴巴集团旗下阿里国际数字商业集团展位的工作人员介绍，不久前，浙江义乌的一家外贸企 业通过阿里国际站的AI服务拿下价值2000万美元的外贸大单。该订单中的沙特阿拉伯客户第一次上线 询盘是凌晨1点多，当时所有外贸业务员都已下班，只有AI生意助手自动接待在线。AI与客户的五、六 轮对话中，不仅询问了买家采购设备的用途、物流接收地等信息，还询问了买家此次采购是为了定期维 护、紧急更换还是特定项目。 从本届数贸会上参展企业的案例和专家观点可以看出，当下的AI技术已可助力跨境贸易 ...

Core Insights - The company has successfully assisted multiple AI glasses companies in launching their first products [1] - The company has signed contracts with several leading manufacturers, securing a dominant market share among top players [1] - Related product solutions have achieved mass production and implementation in various benchmark projects [1]

Group 1 - The company has successfully assisted multiple AI glasses companies in launching their first products [1] - The company has signed contracts with several leading manufacturers, securing a dominant market share among top players [1] - Related product solutions have been mass-produced and implemented in various benchmark projects [1]