报告导读： 上周（ 2025/8/4-2025/8/8 ）机器人领域动态：大疆首款旗舰扫地机器人 发布，上海市人民政府办公厅印发《上海市具身智能产业发展实施方案》， 2025 世界机 器人大会于北京正式开幕。 投资建议： 重点关注机器人整机厂商及机器人核心零部件供应商 ， 包括 ： 1 ）执行器和电机； 2 ）减速器； 3 ）丝杠类； 4 ）丝杠设备类； 5 ）轴承； 6 ）传感器； 7 ）整机。 报告来源 以上内容节选自国泰海通证券已发布的证券研究报告。 报告名称： 上海助推具身智能产业集群发展，多款机器人产品亮相2025世界机器人大会；报告日期： 2025.08.10 报告作者： 肖群稀（分析师），登记编号：S0880522120001 黄龙（分析师），登记编号：S0880525070027 李启文（分析师），登记编号：S0880524060001 重要提醒 本订阅号所载内容仅面向国泰海通证券研究服务签约客户。因本资料暂时无法设置访问限制，根据《证 券期货投资者适当性管理办法》的要求，若您并非国泰海通证券研究服务签约客户，为保证服务质量、 控制投资风险，还请取消关注，请勿订阅、接收或使用本订阅号中的任何信 ...

Core Viewpoint - The article maintains a tactical overweight view on A-shares and US stocks for August, driven by improving market risk appetite and expectations of economic growth [1][2]. Group 1: Market Conditions - Recent improvements in market risk appetite have led to risk assets significantly outperforming safe-haven assets, with equities outperforming commodities and bonds [1]. - The expectation of continued economic growth and government support for capital markets is leading to a highly optimistic outlook for A-shares [1]. - The US stock market is viewed positively due to a reduced probability of recession and increasing chances of interest rate cuts [1]. Group 2: Asset Allocation Strategy - The current low-interest-rate environment necessitates higher demands for asset allocation research, with a long-term trend of increased institutional investment in capital markets [1]. - A strategic asset allocation (SAA) plan has been constructed based on macro factors, achieving an annualized return of 9.1% and a Sharpe ratio of 1.57 as of the end of July [1]. - The proposed strategic benchmark allocation is set at 45% for equities, 45% for bonds, and 10% for commodities, with a deviation limit of 10% [1]. Group 3: Tactical Asset Allocation (TAA) - The improvement in risk appetite is identified as a core factor influencing the current tactical asset allocation, with multiple factors expected to support continued strong performance in equity assets [2]. - The TAA plan projects an annualized return of 55% by 2025, with a high Sharpe ratio of 1.65 based on full sample backtesting [2]. Group 4: Specific Allocation Recommendations - For August, the recommended allocation is 55% in equities, 40% in bonds, and 5% in commodities [3]. - A positive outlook is maintained for A-shares and Hong Kong stocks due to stable economic growth expectations and government support [3]. - Caution is advised regarding government bonds due to pressures from market risk appetite and capital redemption [3]. - The outlook for commodities, particularly oil, remains cautious due to supply and demand pressures [3].

报告导读： GPT-5 发布会展现了 OpenAI 在模型智能性、推理效率与产品体系上的全面 进化，标志着大模型从"强大"走向"可控、高效、可用"的关键拐点。 OpenAI 正式发布其新一 代通用人工智能模型 GPT-5 ，相较此前各代模型在智能水平上实现了实质性飞跃。看好后续 AI 应用层爆发，间接反哺底层硬件， 维持行业"增持"评级。 GPT-5 的一个大的升级是 GPT-5 被设计为一个统一的多模型系统 。具体而言，系统由三大核心组成：一是主模型，负责处理大多数日常查询，具备高效响 应能力；二是" GPT-5 思考"（ GPT-5 Think ）模块，专为复杂问题设计，具备更强的推理与逻辑能力；三是实时模型路由器，能够根据任务的类型、难 度、工具调用需求及用户的显式提示判断并分配最合适的模型执行任务。 GPT-5 实现"复利型体验"提升，用户短期感知不一定强烈，但长时间使用相信会觉得顺滑、稳定、准确度都在提升。 GPT-5 在多项核心能力上实现显著提 升，体现出其作为下一代通用大模型的全面进化。首先，在编程能力方面， GPT-5 在多个权威评测中超越同代模型，如在 SWE-bench Verified ...

Core Viewpoint - The AI glasses market is poised for explosive growth, driven by lightweight design, essential use cases, and consumer pricing, with a projected 188% increase in sales in China by 2025, potentially creating a trillion-yuan market space [1][2]. Market Dynamics - The evolution of AI glasses is transitioning from separate functional devices to integrated computing solutions, with three main product categories: audio devices, audio + camera devices, and AR integration devices [2]. - Major brands like Meta and Xiaomi are leading the consumer market, while domestic players are catching up in technology and application expansion [2]. Value Chain and Market Estimation - The cost structure of AI glasses shows that the main chip accounts for approximately 33% of total costs, with optical display and computing units making up 43% and 31% respectively [3]. - The global market for AI glasses is expected to exceed 190 billion RMB by 2030, with the Chinese market projected to surpass 50 billion RMB, reflecting a CAGR of over 80% over five years [3]. Industry Structure - The AI glasses industry value chain includes upstream manufacturers of optical modules, chips, and frames; midstream ODM/OEM manufacturers and software suppliers; and downstream brand manufacturers and distributors [3].

Core Viewpoint - The article outlines a series of in-depth research calls focused on various aspects of robotics and related technologies, highlighting investment opportunities and industry trends in the mechanical sector [4][5]. Group 1: Robotics and Technology - The research series includes discussions on humanoid robots, their cognitive capabilities, and sensory technologies such as 3D vision and tactile sensors [4]. - Key topics also cover the mechanical components of robots, including efficient motors, precision reducers, and bearings, which are essential for enhancing robotic functionality [4]. - The series emphasizes the growing market for exoskeleton robots and companion robots, indicating a shift towards advanced wearable technology and AI companionship [4]. Group 2: Industrial Applications - The research addresses the logistics sector, focusing on how robotics can enhance supply chain efficiency and the demand for specialized robots in safety and unique operational scenarios [5]. - The article discusses the commercial aerospace sector, highlighting the benefits of high-density rocket launches and the interdependence of rocket and satellite supply chains [5]. - It also explores the potential of controlled nuclear fusion and its impact on capital expenditure in upstream sectors driven by downstream demand [5]. Group 3: Equipment and Machinery - The article reviews the recovery expectations in the tool industry, suggesting a resilient market outlook [5]. - It analyzes the engineering machinery sector, discussing the evolution of domestic sales over the past thirty years and the future trends in electric and unmanned applications [5]. - The research highlights the upcoming boom in mining automation, indicating a shift from technical feasibility to commercial viability [5].

Group 1: Market Strategy and Asset Allocation - The core viewpoint emphasizes an improvement in risk appetite supporting global equity allocation value, with a tactical overweight on A-shares and US stocks as of August [3] - The current low interest rate environment necessitates higher demands for asset allocation research, with a strategic asset allocation (SAA) plan yielding an annualized return of 9.1% and a Sharpe ratio of 1.57 as of the end of July [3] - The proposed strategic benchmark allocation is set at 45% for equities, 45% for bonds, and 10% for commodities, with a deviation limit of 10% [3] Group 2: Tactical Asset Allocation Insights - The enhancement of risk appetite is identified as a key factor influencing current tactical asset allocation, with expectations of continued strong performance in equity assets supported by various factors including technological breakthroughs and government support for capital markets [4] - The tactical asset allocation (TAA) plan anticipates a 55% weight in equities, 40% in bonds, and 5% in commodities, reflecting optimism for A-shares and Hong Kong stocks, as well as marginal optimism for US and Japanese stocks [4] - Caution is advised regarding government bonds due to market risk appetite adjustments and redemption pressures, while commodity prices, particularly oil, may face dual pressures from supply and demand [4] Group 3: Infrastructure Opportunities in Xinjiang - Historical reviews of three central work conferences on Xinjiang indicate significant stock price increases for key construction companies following these meetings, with notable gains such as 45.6% for Beixin Road and Bridge after the first conference [10] - Fixed asset investment in Xinjiang increased by 16.2% in the first five months of the year, with industrial investment rising by 22.8%, particularly in wind and solar energy projects [11] - The establishment of the Xinjiang Tibet Railway Company with a registered capital of 950 billion yuan highlights ongoing infrastructure investment, with plans for 500 key projects totaling 3.47 trillion yuan by 2025 [12] Group 4: Coal Chemical Industry Development - The coal chemical sector in Xinjiang is projected to experience a construction peak, with planned projects exceeding 800 billion yuan, driven by policy support [13] - Companies like China Chemical and Donghua Technology are positioned to capitalize on traditional coal chemical market opportunities, with significant new orders reported [13] Group 5: Aerospace and Satellite Deployment - The acceleration of low Earth orbit (LEO) satellite deployment is identified as a critical area, with a projected need for 2.3 million satellites by 2030, necessitating enhanced rocket launch capabilities [18] - Current rocket launch capacity is insufficient to meet the demand for satellite deployment, with an annual requirement of 1,500-2,000 tons compared to the current capacity of approximately 200 tons [19] - Reusable rocket technology is highlighted as a key factor in reducing launch costs, with liquid fuel rockets becoming the mainstream choice for new generation reusable rockets [20]

Core Insights - TSMC's 2nm technology leak has resulted in the dismissal of multiple employees, raising concerns about the company's reputation and trust [2] - OpenAI has officially released GPT-5, showcasing significant advancements in various fields, which is expected to boost capital expenditure in AI infrastructure [3] - Elon Musk has halted the Dojo chip project at Tesla, indicating a strategic shift towards reliance on traditional computing chips [4] Group 1: TSMC Incident - TSMC confirmed on August 5 that it has dismissed several employees due to the leak of 2nm process trade secrets, involving around 10 individuals [2] - One individual, a former employee who moved to Tokyo Electron, has been arrested for obtaining secrets through connections with advanced process R&D personnel [2] - The incident has caused significant reputational damage and a trust crisis for TSMC, necessitating resource integration and confidence rebuilding to meet production targets [2] Group 2: OpenAI Developments - OpenAI officially launched GPT-5 on August 7, which surpasses previous models in coding, mathematics, writing, health, and visual perception [3] - GPT-5 can create aesthetically pleasing websites and applications with a single prompt and excels in debugging large codebases [3] - The release of two new open-source models, gpt-oss-120b and gpt-oss-20b, marks OpenAI's return to the open ecosystem after six years, reinforcing market expectations for AI model applications [3] Group 3: Tesla's Strategic Shift - Elon Musk ordered the closure of the Dojo supercomputer project on August 7, which was intended for training AI models for autonomous driving and robotics [4] - This decision reflects a deep adjustment in Tesla's AI computing strategy, as evidenced by the mention of new NVIDIA H200 clusters in the company's Q2 2025 financial report [4] - The termination of the Dojo project highlights the challenges of custom chip development, prompting AI companies to increase investments in traditional computing chips [4]

Group 1 - The article discusses the current development stage of the global AI industry, highlighting its rapid growth and transformative potential across various sectors [5][2]. - It emphasizes the advancements in large model development, indicating significant progress in AI capabilities and applications [6][2]. - The article outlines the investment logic behind the AI industry, suggesting that the ongoing innovations present substantial opportunities for investors [2][5]. Group 2 - The article provides insights into the trends within the AI industry, noting the increasing integration of AI technologies in traditional industries [5][2]. - It mentions the competitive landscape, where companies are racing to develop and deploy AI solutions, leading to a dynamic market environment [5][2]. - The article also touches on the regulatory considerations that may impact the growth and adoption of AI technologies in the future [5][2].

Core Viewpoint - The market is expected to maintain a trend of oscillating upward in the coming week, with a notable presence of both bullish and bearish sentiments [1][2]. Market Indicators - The liquidity shock indicator for the CSI 300 index was 2.49, indicating current market liquidity is 2.49 standard deviations above the average level of the past year [2]. - The PUT-CALL ratio for the SSE 50 ETF decreased to 0.92, reflecting a reduced caution among investors regarding the short-term performance of the ETF [2]. - The five-day average turnover rates for the SSE Composite Index and Wind All A Index were 1.06% and 1.65%, respectively, indicating a decline in trading activity [2]. Macroeconomic Factors - The onshore and offshore RMB exchange rates experienced weekly increases of 0.39% and 0.05%, respectively [2]. - In July, China's CPI was reported at 0.0% year-on-year, slightly below the previous value of 0.1% but above the consensus expectation of -0.12%. The PPI remained at -3.6%, matching the previous value and below the consensus expectation of -3.44% [2]. Technical Analysis - The SAR indicator for the Wind All A Index showed an upward breakout on August 6, indicating a potential trend reversal [2]. - The market score based on the moving average strength index is currently at 246, placing it in the 91.5 percentile for 2023 [2]. - The sentiment model score is 3 out of 5, with both trend and weighted models signaling a positive outlook [2]. Market Performance - For the week of August 4-8, the SSE 50 Index rose by 1.27%, the CSI 300 Index increased by 1.23%, the CSI 500 Index grew by 1.78%, and the ChiNext Index saw a rise of 0.49% [3]. - The overall market PE (TTM) stands at 20.7 times, which is in the 67.9 percentile since 2005 [3]. Factor Crowding - The crowding degree for small-cap factors has decreased, with small-cap factor crowding at 0.79, low valuation factor crowding at 0.11, high profitability factor crowding at -0.25, and high growth factor crowding at 0.25 [3]. Industry Crowding - The industries with relatively high crowding degrees include machinery, defense and military, non-ferrous metals, comprehensive, and steel, with notable increases in crowding for defense and machinery sectors [4].

Core Insights - The article highlights the ongoing support of various consumer subsidy policies for durable goods prices, contributing to a sustained increase in core CPI year-on-year, while noting the sluggish recovery in rental and household service prices [1][2]. CPI Analysis - In July, the CPI increased by 0.4% month-on-month, aligning with seasonal trends, while year-on-year figures remained flat. Core CPI rose to 0.8%, the highest since March 2024 [2]. - Key drivers of core CPI include: 1. Strong price increases in consumer policy-supported sectors, with living goods and services prices rising by 0.8% month-on-month, significantly above the seasonal average of 0.26% [2]. 2. Rising gold prices and a surge in consumer gold purchases led to high growth in other goods and services CPI [2]. 3. Summer travel remains robust, although the growth rate of tourism CPI has slightly decreased compared to 2023 and 2024, indicating potential in the tourism and service consumption markets that requires further policy guidance [2]. 4. Rental and household service CPI saw a modest increase of 0.1% month-on-month, which is below historical averages, indicating limited recovery progress [2]. PPI Analysis - In July, the PPI decreased by 0.2% month-on-month and 3.6% year-on-year, with the decline rate remaining stable. The PPI is showing signs of stabilization, supported by raw material prices due to intensified "anti-involution" policies, although downstream factory prices remain weak due to overcapacity and export competition [3]. - The positive impact of "anti-involution" policies on PPI is expected to be gradual and long-term. Current capacity optimization policies have expanded from traditional industries like coal, steel, and cement to emerging sectors such as automotive, photovoltaics, and batteries, affecting nearly 20% of industry revenue [3].