1. 前一天摘要 – 2025年11月24日，星期一 全球股市周一上涨，得益于对美联储近期降息的信心回升以及风险厌恶的普遍放宽。欧洲表现适度上 涨，科技和增长板块主导，收益率下降改善了市场情绪。亚洲市场也上涨，香港及半导体相关股股表现 优于对持续人工智能投资和全球更有利货币环境的期待。尽管地缘政治和估值担忧依然存在，当天的行 动表明全球风险承担将谨慎但明确地回归。 顶级股票搬运者 1. 今天的动机——2025年11月25日，星期二 波动率与收益率 商品（结局/纽约晚期） 国际 美国主要指数 周一，随着美联储降息的乐观情绪再次飙升，美国股市上涨，投资者倾向于超卖的科技股。纽约联储行 长约翰·威廉姆斯及其他联储官员发出鸽派信号，信心被提振，这大幅推高了12月降息的可能性。债券 收益率下降减轻了增长股的负担，而大型科技股，尤其是与人工智能相关股票如Alphabet和特斯拉的强 劲反弹，助推了上涨。技术买家介入，尤其是在巨头股领域，增长股的轮转也重新出现。不过，这次反 弹更像是一次缓解而非彻底重置，因为AI估值担忧依然存在，许多投资者密切关注宏观或政策动荡的 迹象。 标普500指数：▲1.5467%，至6705.12 ...

Core Viewpoint - The article discusses the release of xAI's latest model, Grok 4.1, which has achieved the top position in the text capability rankings, showcasing advancements in dialogue intelligence, emotional understanding, and practical applications in the real world [3][5]. Model Performance - Grok 4.1 Thinking leads the text capability rankings with an Elo score of 1483, while the non-reasoning mode ranks second with a score of 1465 [5]. - The model has a 64.78% probability of being preferred by users compared to its predecessor during blind testing [5]. - Grok 4.1 has significantly improved emotional intelligence, aligning with the recent updates from OpenAI's GPT-5.1, which also focuses on creating more "human-like" interactions [5][6]. Emotional Intelligence - xAI tested Grok 4.1 on the EQ-Bench3, where it ranked first in both reasoning and non-reasoning modes, demonstrating enhanced emotional intelligence capabilities [6]. - The model's responses to emotional prompts are more nuanced and empathetic compared to previous versions, as illustrated by its improved replies to users expressing feelings of loss [6][7]. Creative Writing and Expression - Grok 4.1 shows significant improvements in creative writing, producing more engaging and dramatic narratives compared to its predecessor [8]. - The model's ability to express complex thoughts and emotions has been enhanced, allowing for richer storytelling [8]. Reduction of Hallucinations - The hallucination rate of Grok 4.1 has decreased from 12.09% to 4.22%, indicating a nearly threefold reduction in factual inaccuracies during information retrieval prompts [9]. Development Methodology - xAI utilized a large-scale reinforcement learning infrastructure to optimize the model's style, personality, practicality, and consistency [9]. - New methods were developed to leverage advanced reasoning models as reward models, enabling large-scale self-evaluation and iterative output improvements [9]. Competitive Landscape - The competition in the large model arena is intensifying, with OpenAI and Google also updating their products, raising questions about whether Grok 4.1 will maintain its leading position [9].

盘前要闻 1、外交部、文旅部、教育部等多部门重要提醒！近期避免前往日本 类型：宏观 情绪影响：中性 外交部发布出行提醒，日本领导人公然发表涉台露骨挑衅言论，严重恶化中日人员交流氛围，给在日中 国公民人身和生命安全带来重大风险。文化和旅游部郑重提醒中国游客近期避免前往日本旅游。教育部 建议中国公民谨慎规划赴日留学安排。 2、华为将发布AI领域突破性技术，有望解决算力资源利用效率难题 类型：行业 情绪影响：正面 4、证监会主席吴清：着力推动市场更具韧性、更加稳健，制度更具吸引力 类型：宏观 情绪影响：正面 近日，中国证监会举办学习贯彻党的二十届四中全会精神宣讲报告会，吴清强调，着力推动市场更具韧 性、更加稳健，制度更加包容、更具吸引力，上市公司质量更高、价值更优，监管执法更加有效、更具 震慑力，开放迈向更深层次、更高水平。 华为将在11月21日发布AI领域的突破性技术，可将GPU、NPU等算力资源的利用率，从行业平均的30% 至40%提升至70%，显著释放算力硬件潜能。据透露，华为是通过软件创新实现英伟达、昇腾及其他三 方算力的统一资源管理与利用，屏蔽算力硬件差异，为AI训练推理提供更高效的资源支撑。 3、国常会 ...

Group 1 - Apple plans to enhance its satellite services with new features such as satellite-based mapping and connectivity for third-party applications [1] - Current satellite services for iPhone 14 and newer models, as well as the new Apple Watch Ultra 3, include emergency SOS, messaging, location sharing, and roadside assistance [1] - Apple is utilizing a network operated by Globalstar for its satellite communication services and is exploring the development of its own satellite system [1] Group 2 - New satellite features under development include a framework for developers to integrate satellite connectivity into their applications, enhancing Apple Maps, expanding messaging options, indoor satellite usage, and 5G non-terrestrial network capabilities [1] - The anticipated 5G NTN is expected to debut in the next iPhone series [1] - Apple plans to invest approximately $1 billion annually to utilize a customized Google Gemini model to improve its Siri voice assistant, which is significantly lower in cost compared to its existing Apple Intelligence model [1] Group 3 - Apple is expected to launch new MacBook Pro models (M6 Pro and M6 Max) and a lower-priced MacBook product (J700) by 2026 to compete with more affordable computers [2]

Core Viewpoint - The article discusses the intersection of AI and energy, highlighting the current electricity shortage faced by the AI industry and the innovative solutions being proposed, such as Google's "Project Suncatcher" which aims to utilize solar energy in space for AI computations [5][6][7]. Group 1: AI Industry's Energy Needs - Microsoft CEO Satya Nadella indicated that the AI industry is experiencing a power shortage due to high electricity demands from GPUs, which are not being met by current energy supply [5][17]. - OpenAI CEO Sam Altman suggests that while there is a short-term electricity shortage, the long-term outlook may improve as AI energy consumption decreases over time [6][19]. - The AI industry's energy consumption is projected to double by 2030, with significant increases in data center power demands, highlighting the urgency of addressing energy supply issues [15][16]. Group 2: Innovative Solutions - Google's "Project Suncatcher" aims to deploy satellites in low Earth orbit to harness solar energy for AI computations, potentially overcoming terrestrial energy limitations [7][9]. - The project plans to launch its first test satellites by early 2027, focusing on direct computation in space rather than transmitting energy back to Earth [9][10]. - Other companies, such as Starcloud and initiatives in China, are also exploring space-based data centers, indicating a growing trend towards utilizing space for energy-intensive AI operations [12][13]. Group 3: Energy Supply Challenges - The energy supply chain faces significant delays, with grid access approvals taking up to five years and transmission line construction taking 10 to 17 years, creating a mismatch with the rapid growth of AI demand [17]. - Despite the apparent need for energy, major energy companies have not seen corresponding stock price increases, suggesting market skepticism about the AI industry's energy crisis [16]. Group 4: Future of Energy Generation - Small Modular Reactors (SMRs) are emerging as a viable solution for providing stable, low-carbon energy, with companies like Google and Microsoft investing in this technology [18]. - The global renewable energy capacity is expected to increase significantly, potentially outpacing the energy demands of AI [17][18]. Group 5: Efficiency Improvements - AI models are becoming more efficient, with significant reductions in energy consumption per unit of intelligence, indicating a potential decrease in future energy needs [19][21]. - New AI chips, such as Meta's Athena X1, show substantial improvements in energy efficiency, which could further alleviate the energy demands of AI systems [22][23]. - Data center energy efficiency is improving, with advancements in cooling technologies and energy management systems leading to lower power usage effectiveness (PUE) [24].

Core Insights - TSMC is shifting its focus towards advanced processes and will gradually outsource some 40-90nm orders to its subsidiary, World Advanced [1][3] - TSMC has announced the suspension of its 6-inch wafer fab in Hsinchu and plans to exit the GaN foundry business within two years, reallocating resources to higher-margin businesses [1][4] Industry Performance - The chemical sector performed well this week, with the Shenwan Chemical Index rising by 3.54%, outperforming the CSI 300 Index by 0.82% [2] - Short-term macro policies are relatively stable, with a focus on three key tracks: growth in terminal industries like robotics and AI, low-valuation cyclical sectors, and actual changes in industry fundamentals such as the lithium battery supply chain [2] Major Events - TSMC's exit from mature processes marks a significant strategic shift, focusing on higher-margin business areas [3][4] - The Dutch government anticipates that Ansem China will soon resume chip supplies, with constructive talks with China [4] - The U.S. government is evaluating military options against Venezuela, including seizing oil fields, leading to a slight increase in WTI crude oil prices [4]

Core Insights - Google is strategically positioning itself across the AI value chain, focusing on the development of its TPU chips, OCS technology, and the Gemini model, which is recognized for its global leadership in capabilities [1] - The company is expected to increase its capital expenditures significantly, with projections for 2025 set between $91 billion and $93 billion, indicating a strong growth trajectory in AI hardware and cloud services [1] Group 1: Google’s AI Strategy - Google has established a comprehensive ecosystem involving chips (TPU), networks (OCS), models (Gemini), and applications (cloud computing/search/advertising), which creates a competitive moat [1] - The Gemini model has surpassed 650 million monthly active users, with token processing volume increasing over 20 times within a year [1] - Google Cloud revenue is on the rise, driven by strong demand for AI, with a robust order backlog [1] Group 2: OCS Technology Advantages - OCS (Optical Circuit Switching) technology is crucial for addressing the dynamic computing needs of AI data centers, offering significant advantages in latency and energy consumption [2] - Traditional data center architectures face challenges in power consumption and cost, while OCS provides a solution through server decoupling and efficient data transmission [2] Group 3: TPU and OCS Developments - Google has upgraded its expectations for TPU shipments, with the introduction of OCS technology leading to a 30% increase in network throughput and a 40% reduction in power consumption [3] - The upcoming TPU V7 Ironwood will feature 1.6T optical modules, with OCS technology primarily utilizing MEMS and liquid crystal solutions [3] - Major suppliers involved in OCS components are expected to benefit from this technological advancement, as OCS solutions continue to evolve [3]

Core Insights - The report highlights Google's comprehensive AI stack advantage, which includes chips (TPU), networks (OCS), models (Gemini), and applications (cloud computing/search/advertising), indicating a significant potential for capital expenditure increase [3] - Google's capital expenditure for 2025 has been raised to between 91 billion and 93 billion, with expectations for continued growth in 2026, reflecting a strong demand for AI capabilities and hardware innovation [3] - The Gemini model has surpassed 650 million monthly active users, with token processing volume increasing over 20 times within a year, showcasing robust growth in Google's cloud revenue and order backlog driven by AI [3] Summary by Sections AI Data Center and OCS Technology - OCS (Optical Circuit Switching) technology is crucial for AI data centers, providing advantages such as reduced latency and energy consumption by directly transmitting data using optical signals, which is essential for large-scale AI computing needs [4] - Traditional data center architectures face challenges in power consumption and cost, while OCS offers a key solution for dynamic computing demands through server decoupling and efficient long-distance signal transmission [4] Google TPU and OCS Market Outlook - Google's TPU shipments are expected to increase, with the introduction of OCS technology enhancing network throughput by 30%, reducing power consumption by 40%, and significantly decreasing network downtime [5] - The upcoming TPU V7 Ironwood will feature 1.6T optical modules, with various suppliers poised to benefit from the OCS technology, which includes MEMS and liquid crystal solutions [5] Investment Opportunities - The report suggests focusing on investment opportunities within Google's AI computing supply chain and OCS technology innovations, highlighting key companies such as Zhongji Xuchuang, Changxin Bochuang, and Tengjing Technology as potential beneficiaries [6]

Group 1 - The Shanghai Stock Exchange Sci-Tech Innovation Board 100 Index rose by 1.06%, with notable increases in stocks such as Source Technology (up 11.75%) and Peak Technology (up 5.31%) [3] - The Sci-Tech 100 Index ETF (588030) also increased by 1.06%, with a latest price of 1.33 yuan, and has seen a cumulative increase of 0.53% over the past two weeks [3] - The trading volume for the Sci-Tech 100 Index ETF reached 62.81 million yuan, with an average daily trading volume of 385 million yuan over the past year, ranking first among comparable funds [3] Group 2 - SoftBank Group and OpenAI announced the establishment of a joint venture, SB OAI Japan GK, aimed at providing "Crystal Intelligence" solutions to innovate management and operational models for Japanese companies, with a product launch planned for 2026 [4] - UBTECH Robotics won a bid for a humanoid robot data collection project with a procurement amount of 159 million yuan, with total orders for its Walker series humanoid robots exceeding 800 million yuan for the year [4] - The average growth rate for domestic intelligent robotics companies is expected to be between 50%-100% this year, with the development stage of embodied robots compared to the early years of ChatGPT [4] Group 3 - The Sci-Tech 100 Index ETF closely tracks the Shanghai Stock Exchange Sci-Tech Innovation Board 100 Index, which selects 100 medium-sized and liquid securities from the Sci-Tech Innovation Board [5] - As of October 31, 2025, the top ten weighted stocks in the Sci-Tech 100 Index accounted for 25.77% of the index, including companies like Hua Hong Semiconductor and GuoDun Quantum [5] - The latest scale of the Sci-Tech 100 Index ETF reached 5.756 billion yuan, ranking second among comparable funds, with a significant increase of 6 million shares over the past week [4][5]

Group 1: Guizhou Moutai's Share Buyback and Dividend Plan - Guizhou Moutai announced a share buyback plan with a total fund of no less than 1.5 billion yuan and no more than 3 billion yuan, with a maximum buyback price of 1887.63 yuan per share [2] - The company also revealed a mid-term profit distribution plan for 2025, proposing a cash dividend of 23.957 yuan per share, totaling approximately 30 billion yuan [3] Group 2: Apple's AI Partnership with Google - Apple plans to pay approximately 1 billion USD annually to use Google's 1.2 trillion parameter AI model to upgrade its Siri voice assistant [4] - The new version of Siri is expected to launch in spring 2026, utilizing Google's Gemini model for core functionalities [5] Group 3: OpenAI's Growth and User Base - OpenAI announced that it has surpassed 1 million enterprise customers, with over 7 million seats for its ChatGPT for Work service [9][10] - The company reported that ChatGPT has over 800 million active users weekly, indicating a growing acceptance of AI in the market [9] Group 4: IBM's Layoff Announcement - IBM plans to lay off a small portion of its workforce, affecting a low single-digit percentage of its total 270,000 employees, which could mean around 2,700 job losses [11] - The company has previously replaced hundreds of HR roles with AI, leading to a shift in hiring towards more programmers and sales personnel [12] Group 5: Marriage Registration Statistics - The Ministry of Civil Affairs reported that marriage registrations in the first three quarters of 2025 reached 5.152 million pairs, an increase of 405,000 pairs compared to the same period in 2024 [13][14] Group 6: Corporate Leadership Changes - Zhu Lidan has stepped down as the legal representative of Hongsheng Beverage Group, with Zheng Qundai taking over [16][17] Group 7: Economic Relations Between China and the U.S. - The U.S. will suspend additional tariffs on Chinese goods starting November 10, 2025, as part of a significant agreement reached during U.S.-China trade negotiations [19][20] Group 8: ASUS Apology Regarding Game Security - ASUS issued an apology for a video that negatively impacted Tencent's game security system, ACE, and has taken steps to rectify the situation by removing the video and addressing internal management issues [24][25]