Core Viewpoint - XuanTie, a brand under Alibaba's Damo Academy, focuses on RISC-V architecture innovations and open-source ecosystem development, aiming to provide powerful, intelligent, secure, and open computing architectures for the digital age [2]. Group 1: Event Overview - The XuanTie RISC-V Ecological Conference will take place on March 24, 2026, in Shanghai, guided by the Pudong New Area Government and organized by Damo Academy, showcasing the vitality of RISC-V technology [2]. - The conference will feature a "1+2" innovative insight matrix with two sub-forums alongside the main forum, discussing key topics such as high-performance architecture, AI industry applications, and trends in ecological chip products [3]. Group 2: Forum Highlights - The Alibaba Ecosystem Forum will include teams from Alibaba Cloud, Ant Group, and Damo Academy, sharing new cases of RISC-V integration in cloud computing, large models, and intelligent terminals [3]. - The Chip Industry Sub-Forum will invite leading companies like Hongsi Electronics and Beijing Zhixinwei to present insights and practical experiences regarding RISC-V applications across various industries [3]. Group 3: Conference Agenda - The agenda includes keynote speeches from industry leaders, discussions on RISC-V's role in cloud computing, and the unveiling of new partnerships and awards for outstanding contributions to the RISC-V ecosystem [7][8].

互联网大厂正在加速落地AI产品与应用！ 继国内多家互联网巨头推出AI购物、订外卖、订酒店等功能之后，又一重要生活场景迎来AI应用。 继2025年9月，滴滴开启AI出行助手公测后，近日升级推出v1.0版本。这也是行业首个AI打车，意味着所有注册用户均可使用该服务。用户在滴滴出行APP中体 验"AI叫车"，只需要一句话，无论是对车辆有个性化要求，还是自身状态的模糊表述，AI小滴都能帮助用户找到符合其需求的车辆。 券商中国记者更新最新版本APP后经过实测，滴滴此次推出的AI打车功能，可仅一句话精确匹配用户多类型需求：例如不易晕车、多途经点、孕妇用车、车型新旧 等复杂场景。 多位分析师指出，当前AI大模型正进入务实落地期，行业进入精细化运营和商业化验证的新阶段。其中，拥有细分行业数据壁垒与场景理解能力的企业，在开发行 业AI智能体方面更具优势，能够将大模型技术与自身业务深度结合，实现差异化竞争。 记者实测：可实现一句话个性化叫车 在测试复杂用车场景时，券商中国记者发现，该功能可根据用户详细需求实现车辆的精准匹配。 例如，记者用语音输入方式，向AI助手发出"现在从朝阳公园去首都机场3号航站楼，我带着老人和小孩，一定要空 ...

Group 1: Market Outlook - The observation window for potential market recovery is late March, when negative factors may be fully priced in[2] - March marks the peak of share lock-up expirations, with nearly HKD 1 trillion set to be released, significantly higher than the previous year's end[2] - The second quarter will see a notable decline in lock-up expirations, historically leading to a bottoming effect for individual stocks[2] - The annual report disclosures for major tech companies will conclude by the end of March, reducing earnings uncertainty[2] Group 2: External Factors - The postponement of Trump's visit to China from late March to April may improve market sentiment and alleviate valuation pressures[2] - Geopolitical tensions, particularly related to the US-Iran conflict, have led to a downward adjustment of interest rate cut expectations for 2026 to zero[2] Group 3: Investment Strategy - If market sentiment improves in mid to late March, focus on Hang Seng Technology and Hong Kong Internet stocks, as they may benefit from AI application advancements[2] - In the event of unexpected liquidity tightening, consider investing in high-dividend stocks, as historical data shows a 91% probability of the Hong Kong dividend index rising during the dividend season[2] Group 4: Risks - Key risks include geopolitical tensions, overseas inflation risks, and low expectations for domestic growth policies[2] - The current short-selling volume is around 12%, similar to levels seen during the 2021-2022 bear market, indicating limited further downside potential[4]

Core Viewpoint - The article discusses the impact of geopolitical tensions, particularly the escalation of the US-Iran conflict, on global energy prices and the semiconductor materials industry, highlighting the interconnectedness of these sectors and the potential investment opportunities arising from these dynamics [5][6][8]. Group 1: Geopolitical Impact on Energy and Semiconductor Supply Chains - The escalation of the US-Iran conflict has pushed Brent crude oil prices from over $70 to above $100 within a month, affecting global supply chains [5]. - Japan and South Korea, major players in the semiconductor materials market, rely heavily on oil imports from the Middle East, which could disrupt their production capabilities [7]. - Previous geopolitical conflicts, such as the Russia-Ukraine war, have already shown how supply chain disruptions can lead to price surges in critical materials like neon gas, impacting semiconductor manufacturing [9]. Group 2: Recovery in the Semiconductor Materials Industry - Prior to the geopolitical tensions, the semiconductor materials industry was emerging from a downturn, with leading companies reporting significant profit increases for 2025 [12][14]. - Shanghai XinYang reported a 71.12% increase in net profit, with semiconductor business revenue reaching 1.517 billion, a 46.5% year-on-year increase [15]. - The global semiconductor sales are projected to reach $791.7 billion in 2025, a 25.6% increase, with the Chinese market expected to grow by 17.3%, surpassing $200 billion [16]. Group 3: Demand Drivers in the Semiconductor Sector - The demand surge is primarily driven by the explosion of AI applications, leading to increased requirements for high-bandwidth memory (HBM) and AI chips, filling production capacities across storage and logic chips [18]. - The automotive sector is also experiencing a rise in chip usage, with electric vehicles requiring significantly more semiconductors, further driving demand for semiconductor materials [19]. Group 4: Supply Chain Dynamics and Material Demand - The expansion of wafer production is ongoing, with global silicon wafer shipments expected to grow by 5.4% to 12,824 million square inches by 2025 [21]. - The industry has shifted from an oversupply situation to a tight balance between supply and demand, exacerbated by the recent geopolitical tensions [24][25]. Group 5: Investment Logic in Semiconductor Materials - The investment logic in semiconductor materials combines long-term trends of domestic substitution with short-term catalysts from geopolitical events [28]. - The domestic market has made progress in mid-to-low-end materials, but high-end material production remains critically low, indicating significant market opportunities for domestic producers [30]. Group 6: Key Opportunities in Semiconductor Materials - The most vulnerable segment in the current geopolitical context is photolithography materials, where Japanese companies dominate over 90% of the high-end market [44]. - The supply of rare gases, essential for photolithography equipment, is also at risk due to geopolitical tensions, creating opportunities for domestic producers [46]. - The large silicon wafer market, which constitutes nearly one-third of the materials market, is also poised for growth as domestic companies are ready to capitalize on potential price increases and domestic substitution orders [46]. Group 7: Conclusion on Globalization and Supply Chain Security - The article emphasizes that the recent geopolitical uncertainties highlight the importance of supply chain security over mere efficiency, making domestic substitution in semiconductor materials a critical focus for the industry [50][52].

Core Viewpoint - Leap Motor achieved significant milestones in vehicle deliveries and profitability, becoming the second new car manufacturer to report annual profits after Li Auto, with a net profit of 540 million yuan last year [1][6]. Group 1: Financial Performance - Leap Motor's monthly delivery volume exceeded 70,000 units, and it completed its annual sales target of 500,000 units 45 days ahead of schedule [1]. - The company reported a net profit of 540 million yuan, marking a turnaround from a net loss of 2.82 billion yuan in 2024 [6][13]. - Revenue and delivery volume are closely correlated, with revenue increasing from 32.16 billion yuan for nearly 300,000 vehicles sold to 64.73 billion yuan for nearly 600,000 vehicles sold [8]. Group 2: Scale and Cost Control - Leap Motor's gross margin increased significantly, rising from 0.5% to 14.5% over three years, with a record gross margin of 15% achieved in Q4 2025 [10][14]. - The company emphasizes scale as a key to profitability, aiming to replicate the success of brands like Uniqlo by focusing on volume rather than high margins [14]. - Leap Motor has increased the proportion of self-researched and manufactured components to over 65% of the vehicle BOM, with a target of 80% for the current year [21][23]. Group 3: Product Strategy - Leap Motor introduced a diverse product lineup, including the flagship D19 SUV and the A10 compact car, which are expected to drive sales growth [24][25]. - The company plans to enhance its autonomous driving capabilities, with a clear roadmap for the rollout of advanced features by 2026 [32][44]. - Leap Motor's strategy includes not only selling vehicles but also potentially providing technology solutions to other automotive brands, expanding its business model [67]. Group 4: Global Expansion - Leap Motor aims to achieve annual sales of 1 million vehicles by 2026, with a long-term goal of 4 million vehicles in the next decade [47][51]. - The company has entered 35 countries and regions, exporting 67,000 vehicles in 2025, leading among new car brands in exports [58][60]. - Strategic investments from major players like Stellantis and FAW have bolstered Leap Motor's market position and provided opportunities for collaboration in technology development [55][56].

Investment Rating - The investment rating for BOSS Zhipin (BZ.O) is "Outperform the Market" [6][21]. Core Views - In Q4 2025, the company achieved revenue of 2.079 billion yuan, a year-on-year increase of 14%, exceeding management's guidance of 11.4%-13.0% [1][8]. - The adjusted Non-GAAP operating profit reached 900 million yuan, up 36.6% year-on-year, while the Non-GAAP net profit attributable to shareholders was 908 million yuan, a 14.4% increase, also surpassing Bloomberg consensus expectations [1][8]. - Cash collections in Q4 2025 amounted to 2.230 billion yuan, reflecting a year-on-year growth of 21.4% and a sequential increase from Q3 [1][8]. Business Segmentation Summary - The enterprise online recruitment business (B-end) generated revenue of 2.067 billion yuan, a 14.6% increase, with the number of paying enterprise clients reaching 6.8 million, up 11.5% year-on-year [2][10]. - The average revenue per user (ARPU) for the past 12 months was 1,205 yuan, a 1.1% increase year-on-year, indicating steady growth across various customer segments [2][10]. - Other business segments, primarily providing paid value-added services to job seekers, reported revenue of 11.64 million yuan, down 40.3% [2][10]. - The company has introduced AI tools for both B-end and C-end, including "AI Fast Recruitment" and AI-assisted interview features, while continuing to develop its proprietary model for practical business applications [2][10]. Financial Forecasts - For Q1 2026, the company expects revenue to be between 2.05 billion and 2.085 billion yuan, representing a year-on-year growth of 6.6%-8.4% [3][20]. - The company plans to allocate no less than 50% of the previous fiscal year's adjusted net profit for dividends and buybacks over the next three years, with a buyback cap raised to 400 million USD [3][20]. - Adjusted net profit forecasts for 2026 and 2027 are 4.227 billion yuan and 4.797 billion yuan, respectively, with corresponding PE ratios of 10.7 and 9.4 [5][21].

Investment Rating - The investment rating for BOSS Zhipin (BZ.O) is "Outperform the Market" [6][21]. Core Views - In Q4 2025, the company achieved revenue of 2.079 billion yuan, a year-on-year increase of 14%, exceeding management's guidance of 11.4%-13.0% [1][8]. - The adjusted Non-GAAP operating profit reached 900 million yuan, up 36.6% year-on-year, while the Non-GAAP net profit attributable to shareholders was 908 million yuan, a 14.4% increase, also surpassing Bloomberg consensus expectations [1][8]. - Cash collections in Q4 2025 were 2.230 billion yuan, reflecting a year-on-year growth of 21.4% and a sequential increase from Q3 [1][8]. Business Segmentation Summary - The enterprise online recruitment business (B-end) generated revenue of 2.067 billion yuan, a 14.6% increase, with 6.8 million paid enterprise clients over the past 12 months, up 11.5% year-on-year [2][10]. - The average ARPU (Average Revenue Per User) for the past 12 months was 1,205 yuan, a 1.1% increase year-on-year, indicating steady growth across various customer segments [2][10]. - Other business segments, primarily providing paid value-added services to job seekers, reported revenue of 11.64 million yuan, down 40.3% [2][10]. - The company introduced AI tools such as "AI Speed Recruitment" for B-end and AI-assisted interview features for C-end, while continuing to develop its proprietary model for practical business applications [2][10]. Financial Performance and Forecast - The Non-GAAP operating profit margin in Q4 2025 reached 43.3%, an increase of 7.2 percentage points year-on-year, indicating significant efficiency improvements [2][15]. - For Q1 2026, the company expects revenue to be between 2.05 billion and 2.085 billion yuan, representing a year-on-year growth of 6.6%-8.4% [3][20]. - The company plans to allocate no less than 50% of the previous fiscal year's adjusted net profit for dividends and buybacks over the next three years, with a buyback cap raised to 400 million USD [3][20]. - Adjusted net profit forecasts for 2026 and 2027 are 4.227 billion yuan and 4.797 billion yuan, respectively, with corresponding PE ratios of 10.7 and 9.4 [5][21].

Investment Rating - The report maintains a "Neutral" rating for the computer industry and an "Accumulate" rating for Hongsoft Technology (688088) [2][26]. Core Insights - The demand for computing power continues to be released, driven by AI applications like OpenClaw, leading to significant increases in token usage and price hikes in cloud services [12][25]. - Shanghai plans to allocate 1 billion yuan annually for computing power vouchers to establish the largest computing power scheduling platform in China [12]. - The AI application landscape is rapidly evolving, with domestic models like MiniMax M2.5 and Kimi K2.5 expected to become core choices for high token consumption applications due to their cost advantages [25][26]. Industry News - Shanghai is establishing a major computing power scheduling platform with an annual budget of 1 billion yuan to assist enterprises in accessing computing resources [12]. - AI applications such as OpenClaw are driving a surge in token usage, prompting Alibaba Cloud to raise prices by up to 34% for certain products [12][13]. - NVIDIA launched the Vera Rubin platform, significantly reducing the cost per token to one-tenth of previous levels, enhancing efficiency in AI model training and inference [14]. Company Announcements - Xingyun Technology announced that its wholly-owned subsidiary received a bid notification, indicating a significant project that could impact its future operations [16][17]. - Guoao Technology disclosed plans to sell a 23.16% stake in a subsidiary, with the transaction valued at no less than 400 million yuan [18][19]. Market Review - From March 12 to March 18, the CSI 300 index fell by 0.98%, while the Shenwan Computer Industry index decreased by 2.71%, with all sub-sectors experiencing declines [19][20]. - The current price-to-earnings ratio for the Shenwan Computer Industry is 202.04 times, with a valuation premium of 1397.57% compared to the CSI 300 [20]. Weekly Strategy - The report highlights the increasing demand for high-performance GPU cloud services and AI-specific computing resources, driven by the growth of AI applications [25]. - The AI industry is expected to enter a new phase of development in 2026, with multiple catalysts accelerating commercialization and investment opportunities in the computing power supply chain [25][26].

Core Viewpoint - The article discusses the emergence of OpenClaw, an AI software that enhances productivity by enabling direct computer operations, and its application in the automotive industry through the launch of the "Super Intelligent Agent IM Ultra Agent" by Zhiji Auto, which aims to revolutionize user interaction with vehicles [2][6][8]. Group 1: OpenClaw and Its Capabilities - OpenClaw is an open-source AI software that can perform tasks such as organizing files, replying to emails, and making reservations, marking a shift from simple conversational AI to a more functional digital assistant [2][5][6]. - The software's ability to operate computers directly represents a significant advancement in AI tools, transitioning from mere conversation to actionable tasks [6]. Group 2: Zhiji Auto's Super Intelligent Agent - Zhiji Auto introduced the "Super Intelligent Agent IM Ultra Agent," which aims to provide comprehensive assistance by understanding user preferences and executing tasks with minimal input [8][10]. - Unlike traditional smart cabins that only support basic dialogue, the Super Intelligent Agent can handle complex tasks like ordering coffee by understanding user habits and preferences [12][15]. Group 3: Advanced Features and User Interaction - The Super Intelligent Agent can interpret nuanced user statements, such as "the child is asleep," to adjust vehicle settings for a quieter and more comfortable environment [18][21]. - It can also respond to user emotions and tone, enhancing the driving experience by acting as a digital assistant that understands and anticipates needs [22][23]. Group 4: Technical Foundations - The system is built on a new integrated architecture that merges chassis control, intelligent driving AI, and intelligent cabin AI, allowing for a seamless interaction between these components [25][26]. - Zhiji Auto employs a fully controlled chassis system with triple redundancy to ensure safety and reliability, even without mechanical connections [31][34]. Group 5: AI Models and Collaborations - The IM AD ZETA model, developed in collaboration with Momenta, enhances driving capabilities through reinforcement learning, significantly improving performance compared to existing models [38][39]. - The integration of Alibaba's ecosystem allows for seamless data flow between various applications, enabling the Super Intelligent Agent to execute tasks efficiently [45][46]. Group 6: Future Implications and Safety Measures - The introduction of AI in vehicles signifies a pivotal shift towards smarter, more interactive automobiles that serve as intelligent partners in daily life [55][56]. - Zhiji Auto emphasizes the importance of safety and data security, implementing strict controls and a "black box" mechanism to ensure user information is protected [52][53].

Investment Rating - The report maintains a "Buy" rating for Hikvision (002415.SZ) [3][5]. Core Insights - Hikvision showcased its AI capabilities at AWE2026, emphasizing the integration of AI large models into digital operations to address challenges in information retrieval, inspection efficiency, and management complexity [2]. - The company is expected to achieve revenue of CNY 925.52 billion, CNY 982.30 billion, and CNY 1,067.46 billion for the years 2025, 2026, and 2027 respectively, with net profits of CNY 141.01 billion, CNY 152.07 billion, and CNY 162.85 billion during the same period [3]. Financial Summary - Revenue (CNY million) for 2023A, 2024A, 2025E, 2026E, and 2027E is projected to be 89,341, 92,496, 92,552, 98,230, and 106,746 respectively, with year-over-year growth rates of 7.4%, 3.5%, 0.1%, 6.1%, and 8.7% [4]. - Net profit (CNY million) for the same years is expected to be 14,108, 11,977, 14,101, 15,207, and 16,285, with year-over-year growth rates of 9.9%, -15.1%, 17.7%, 7.8%, and 7.1% [4]. - The latest diluted EPS is projected to be CNY 1.54, CNY 1.31, CNY 1.54, CNY 1.66, and CNY 1.78 for the years 2023A to 2027E [4]. - The company’s total market capitalization is approximately CNY 289,793.24 million, with a total share count of 9,164.87 million shares [5].