1.2亿美元！这家公司刷新具身智能行业天使轮融资 纪录 首席科学家丁文超，是具身智能领域青年科学家、复旦大学机器人研究院研究员，曾入选华为"天 才少年"项目，曾从0到1主导华为智驾端到端决策网络，打造复旦大学首个人形机器人。 该公司首席架构师陈同庆，是原华为ADS智能导航部部长，空间感知首席技术专家；其首席战略官 Vincent，曾任华为、百度关键技术岗位负责人。 天眼查数据显示，今年3月6日，它石智航的注册资本发生变更，由原本的100万元增至约136万元。 与此同时，高瓴、启明创投、蓝驰创投、线性资本、联想创投、襄禾资本、恒旭资本等知名投资机构亦 成为其股东。 当前，具身智能领域正在加速发展。从市场规模来看，据麦肯锡预测，到2030年，全球具身智能市 场规模可能达数十万亿元。 今年的《政府工作报告》提出，建立未来产业投入增长机制，培育生物制造、量子科技、具身智 能、6G等未来产业。这也是"具身智能"首次被写入《政府工作报告》中。 3月26日，具身智能初创公司它石智航（TARS）宣布完成天使轮1.2亿美元融资。 本轮融资由蓝驰创投、启明创投共同领投，线性资本、恒旭资本、洪泰基金、联想创投、襄禾资 本、高瓴创投 ...

Investment Rating - The report maintains a "Buy" rating for the company with a target price of 68.73 HKD [7][24]. Core Insights - The company's 4Q revenue increased by 8.7% year-on-year to 35.38 billion RMB, aligning with consensus expectations of 35.7 billion RMB. The gross margin improved by 0.9 percentage points to 54%, and adjusted net profit rose by 13.3% year-on-year to 4.7 billion RMB, also meeting expectations [1][19]. - For 2025, total revenue is expected to grow by 11% year-on-year, with advertising revenue and GMV growth rates slowing to 14% and 13%, respectively. Adjusted net profit is projected to reach 20.6 billion RMB [1][20]. - The company is anticipated to accelerate revenue growth starting in 2Q 2025, driven by AI commercialization opportunities in areas such as 2C subscriptions, 2B e-commerce advertising, and API calls, which could contribute an additional 400 to 800 million RMB in revenue [1][20]. Revenue and Business Segments - E-commerce and other revenue growth slowed to 14.1% year-on-year in 4Q, below expectations by 3.5%. The e-commerce GMV grew by 14.4% year-on-year, with the share of general merchandise e-commerce increasing to 30% [2]. - The number of active e-commerce merchants increased by over 25% year-on-year, with significant growth in GMV from small and medium-sized businesses driven by new policies [2]. - The AI capabilities of the company are expected to enhance advertising efficiency, with a 13.3% year-on-year increase in advertising revenue in 4Q, although this was 2% below expectations [3]. Financial Forecasts and Valuation - The revenue forecast for 2025 has been adjusted upward by 0.6%, while the adjusted net profit forecast has been reduced by 12.6% to 20.6 billion RMB due to increased investments in AI [20][22]. - The report introduces a new valuation for 2027, projecting revenue and adjusted net profit of 163.8 billion RMB and 28.3 billion RMB, respectively [20]. - The target price of 68.73 HKD is based on a sum-of-the-parts (SOTP) valuation, with the advertising business valued at 53.36 HKD per share, the e-commerce business at 8.75 HKD, and the live streaming business at 4.87 HKD [24][25].

Group 1 - AI computing chips are considered the "engine of the AI era," with significant growth in the Sci-Tech Innovation Board ETF Huaxia (589000) experiencing net inflows for four consecutive days [4][3] - As of March 26, 2025, the Sci-Tech Innovation Board Composite Index (000680) increased by 0.38%, with notable stock performances including Tailin Microelectronics (688591) up 16.77% and Airo Energy (688717) up 8.53% [1][4] - The Sci-Tech Innovation Board ETF Huaxia has seen a scale increase of 47.32 million yuan in the past week, ranking first among comparable funds [3] Group 2 - The global computing power scale is expected to grow significantly, from 1397 EFLOPS in 2023 to 16 ZFLOPS by 2030, with a compound annual growth rate of 50% from 2023 to 2030 [4] - AI servers are identified as the core infrastructure supporting generative AI applications, with AI computing chips providing the foundational support for computing power [4] - The top ten weighted stocks in the Sci-Tech Innovation Board Composite Index account for 23.9% of the index, with companies like Haiguang Information (688041) and Cambricon (688256) among the leaders [5]

隔夜美股小幅收涨，中概指数跌1.22%，港股互联 网ETF(159568)早盘走高 东北证券指出，尽管港股市场短期面临震荡，但长期趋势不变。短期内，南下资金流入速度有所放 缓，且面临财报压力和中美摩擦的冲击。长期来看，互联网大厂的资本开支和AI应用的落地将带来生 态变化，险资对红利的偏好也使得港股具有较高的性价比。 隔夜美股小幅收涨，中概指数跌1.22%。昨日大幅下跌的港股三大指数高开（2025年3月26日）， 恒指涨0.5%，国指涨0.31%，恒生科技指数涨0.36%。盘面上，连续下跌的大型科技股多数上涨，阿里 巴巴、网易涨0.7%，百度、美团、小米飘红，快手、京东微幅下跌。 港股互联网ETF（159568）凭借高成长性、强弹性及龙头集中度，成为布局港股科技板块的高效工 具。但其高波动性及政策敏感性需投资者结合自身风险承受能力谨慎参与。若看好AI技术落地及港股 估值修复，可将其作为卫星配置，平衡组合收益与风险。 相关ETF方面，港股互联网ETF（159568）盘中持续震荡，上涨超1%，持续溢价，溢价率超1%， 买盘强烈，成交额近4000万，交投活跃。成分股中多数上涨，金蝶国际涨近5%；阿里影业、第四范式 涨 ...

Investment Rating - The report maintains a "Buy" rating for the company with a target price of HKD 68.73 [7][24]. Core Insights - The company's Q4 revenue grew by 8.7% year-on-year to RMB 35.38 billion, aligning with consensus expectations of RMB 35.7 billion. The gross margin improved by 0.9 percentage points to 54%, and adjusted net profit increased by 13.3% year-on-year to RMB 4.7 billion, also meeting expectations [1][19]. - For 2025, total revenue is expected to rise by 11%, with advertising revenue and GMV growth slowing to 14% and 13% respectively, while adjusted net profit is projected to reach RMB 20.6 billion [1][20]. - The company is anticipated to accelerate revenue growth starting in Q2 2025, driven by AI commercialization opportunities in areas such as 2C subscriptions, 2B e-commerce advertising, and API calls, which could contribute an additional RMB 400-800 million in revenue [1][20]. Revenue and Profitability - Q4 e-commerce and other revenue growth slowed to 14.1%, below expectations by 3.5%. E-commerce GMV grew by 14.4% year-on-year, with a notable increase in the number of active merchants [2]. - The AI capabilities are expected to enhance advertising efficiency and optimize targeting, with Q4 advertising revenue growing by 13.3% year-on-year [3][14]. - The company has adjusted its revenue forecasts for 2025 and 2026, with a slight increase of 0.6% for 2025 and a minor decrease of 0.1% for 2026, while the adjusted net profit forecast for 2025 is revised down by 12.6% to RMB 20.6 billion [20][22]. Valuation and Market Position - The report introduces a new valuation for 2027, projecting revenue and adjusted net profit of RMB 163.8 billion and RMB 28.3 billion respectively. The target price has been raised to HKD 68.73 based on a sum-of-the-parts (SOTP) valuation approach [4][24]. - The valuation reflects a price-to-earnings (PE) ratio of 10.2 times for the advertising business, which is at a discount compared to comparable companies, indicating potential for growth as advertising efficiency improves with AI [24][25]. Business Segments - The "live streaming+" model continues to innovate, driving growth in traditional sectors, with significant increases in user engagement metrics for related services [12]. - The local lifestyle business saw GMV double year-on-year, with a 52.4% increase in monthly paying users, indicating strong demand and effective service optimization [13]. - The AI tool "可灵" has shown promising results in enhancing content creation and advertising efficiency, with significant revenue contributions expected from its commercialization efforts [14].

Core Viewpoint - Shulex, a cross-border e-commerce AI Agent company, has recently completed a financing round of over 100 million yuan, led by Shengda Capital with participation from Northern Light Venture Capital and Starting Gate Fund [1] Group 1: Company Overview - Shulex is backed by Hangzhou Shulihangjian Technology Co., Ltd., which was established in 2021 and focuses on global AI SaaS solutions for the cross-border e-commerce sector [1] - The company was founded by Guo Chenlu, a former core member of DingTalk, who has over 14 years of experience at Alibaba, contributing to projects like Alipay and the development of DingTalk [2] - The team at Shulex consists of professionals from top tech companies such as Alibaba, NetEase, ByteDance, Baidu, and Google [3] Group 2: Product and Innovation - Shulex leverages strong in-house product development capabilities to release a series of AI tools that aim to disrupt traditional practices in the cross-border e-commerce industry [4]

Core Viewpoint - The competition in the large model sector has entered a new phase, with a shift from competition to collaboration among major players, emphasizing the importance of openness and user value in the AI landscape [2][5][36]. Group 1: Industry Dynamics - In 2023, the large model market saw intense competition, with Baidu launching the Wenxiao Yan model 3.5, leading to a frenzy among manufacturers to enhance foundational model technology [2]. - By 2024, the focus shifted to application, resulting in a "bone fracture" price war in the ToB market and a "money-splashing" user acquisition battle in the ToC sector [2]. - The entry of Deepseek as a disruptive player has prompted existing companies to rethink their strategies, leading to a trend of collaboration rather than pure competition [5][8]. Group 2: Product Development and Strategy - Deepseek's emergence has led to a reevaluation among AI manufacturers, with many recognizing the necessity of true openness and collaboration to survive [5][6]. - Baidu's Wenxiao Yan has adopted an open approach, integrating with Deepseek and enhancing its product ecosystem, which has allowed it to maintain competitiveness despite the challenges posed by new entrants [7][21]. - The integration of multiple models, including Deepseek and Baidu's latest models, allows Wenxiao Yan to offer comprehensive services, enhancing user experience through multi-modal capabilities [11][12][31]. Group 3: User-Centric Approach - The AI industry in 2025 will face significant challenges, necessitating new methods to address evolving user needs [33]. - Respecting user value is crucial, as it involves understanding and meeting diverse user demands, which has led to a trend of embracing open-source ecosystems [35][36]. - Baidu plans to make Wenxiao Yan fully free, providing advanced features to users, reflecting a commitment to user-centric development in the competitive landscape [36].

Core Viewpoint - Tencent Music (NYSE: TME) continues to show a trend of fundamental divergence, with strong growth in online music services offset by weakness in social entertainment, leading to a cautious outlook on the stock [1] Group 1: Financial Performance - Music subscription revenue increased by 18% year-on-year to RMB 4.03 billion, driven by a 13% growth in paid user numbers and gradual ARPU improvement [2] - Social entertainment revenue declined by 13% year-on-year to RMB 1.6 billion due to adjustments in live interaction features and stricter compliance procedures [2] - Projected revenue for 2026 is expected to reach RMB 29.33 billion, with a target multiple of 5.0x, resulting in an implied value of approximately USD 13 per share, reflecting a 10% downside from current levels [5][6] Group 2: User Engagement and Content Strategy - Tencent Music is enhancing platform stickiness to support future ARPU growth, expanding its content library through partnerships with SM Entertainment and Kakao Entertainment [3] - The company is investing in application optimization and AI-driven recommendations, resulting in a 10% increase in user playlists and improved sound quality [3] - SVIP membership numbers are steadily increasing, with enhanced features like high-definition online concerts and an expanded digital library attracting more users [3][4] Group 3: Market Outlook and Valuation - Given the structural decline in social entertainment, Tencent Music's valuation should be based solely on its music subscription and advertising business [4] - Future growth expectations include an 11% CAGR in paid users and a 5% CAGR in ARPU over the next two years, with advertising revenue expected to grow at a 15% CAGR [4][6] - The company is also exploring partnerships for in-car music services, which could drive user growth as the automotive market shifts towards high-tech electric vehicles [4]

Core Viewpoint - The article discusses the emergence and implications of the "opening box" industry, which allows individuals to easily access personal information about others, leading to significant privacy concerns and potential abuse [3][4][81]. Group 1: Incident Overview - The article begins with the incident involving a Baidu executive's daughter, clarifying that the data accessed was not from Baidu but from foreign software [3][4]. - It highlights the ease with which personal information can be accessed through the "opening box" service, likening it to an upgraded version of "human flesh search" [5][6]. Group 2: Evolution of Information Access - The transition from an anonymous internet era to one where personal information is easily exposed is discussed, emphasizing the role of social media in this shift [9][10][13]. - The article explains how "human flesh search" initially served as a form of public oversight but has evolved into a tool for online violence due to its accessibility [14][15]. Group 3: The "Opening Box" Industry Chain - The "opening box" industry has developed a complete supply chain, with data leaks primarily originating from social engineering and insider threats [17][18]. - Social engineering is defined as exploiting human weaknesses to gain access to sensitive information, making it difficult to ensure data security [19][20]. Group 4: Data Leakage Mechanisms - The article describes how hackers can easily breach databases and sell the information, with two main methods: spontaneous and customized data extraction [22][25]. - The concept of "credential stuffing" is introduced, where hackers exploit reused passwords across different platforms to gain unauthorized access [29][30]. Group 5: Role of Insiders - Insiders, such as employees with access to sensitive data, pose a significant risk, as they may sell information for personal gain [33][34]. - The article illustrates how the vast and unregulated data systems in China contribute to the ease of data leaks [34][35]. Group 6: Social Engineering Databases - The "social engineering database" serves as a middle layer in the "opening box" industry, where data is aggregated and sold to various entities, including fraudsters [42][45]. - The article notes that these databases allow for the creation of comprehensive profiles on individuals, leading to severe privacy violations [54][55]. Group 7: Market for Personal Information - The primary customers for social engineering databases are often involved in fraudulent activities, as they can purchase large volumes of personal data [60][61]. - The article explains how the ease of accessing personal information has led to a rise in scams and targeted harassment [62][63]. Group 8: Legal and Regulatory Challenges - Existing laws regarding online privacy and data protection are discussed, highlighting the challenges in enforcement and the need for more robust regulations [83][84]. - The article emphasizes that while there are laws against data misuse, the execution and effectiveness of these laws remain questionable [87][88]. Group 9: Future Implications - The article concludes by stressing the urgent need for national attention to the "opening box" industry, likening it to the fight against fraud [98][99]. - It calls for a comprehensive approach to protect individual privacy and dignity in the digital age, as the current trajectory poses significant risks to society [96][97].

Core Viewpoint - The article emphasizes Beijing's pivotal role in China's AI development, highlighting its talent pool, research initiatives, and strategic investments that position it as a leader in the AI industry [6][10][28]. Group 1: Talent and Research - Beijing is recognized as the primary city for AI talent in China, housing 60% of the nation's AI professionals and over 90 prestigious universities and research institutions [27][28]. - The establishment of the Beijing Academy of Artificial Intelligence (BAAI) marked a significant step in uniting top researchers and resources to advance AI research, supported by the local government [19][22]. - The article notes that major AI projects, such as "Wudao," have emerged from BAAI, positioning it as a leading research hub comparable to OpenAI and Google [27][28]. Group 2: Competitive Landscape - The competition among cities for AI talent and companies has intensified, with Beijing offering subsidies and support to lower R&D costs for enterprises [30][40]. - Other cities like Shanghai and Shenzhen are also actively trying to attract AI firms, leading to a "talent war" where companies are concerned about losing key personnel to competitors [31][33]. - Beijing has established an AI industry investment fund of 10 billion yuan, specifically aimed at supporting local AI companies and ensuring they remain in the city [40]. Group 3: Strategic Initiatives - In response to the global AI landscape, Beijing has launched several initiatives to enhance its AI capabilities, including the "Implementation Plan for Accelerating the Construction of a Globally Influential AI Innovation Source" [57][59]. - The city is focusing on developing foundational technologies and innovative solutions to challenge existing models like Transformer, with local companies making significant advancements [57][58]. - Beijing's strategic vision includes becoming a global center for scientific innovation and AI development, aiming to elevate its status on the world stage [46][48]. Group 4: Future Outlook - The article suggests that Beijing's advantages in AI will continue to grow, with top international scholars and companies establishing a presence in the city, further enriching its talent and resource pool [61][63]. - The concentration of major tech firms and research institutions in Beijing is seen as a significant competitive edge, reinforcing its leadership in the AI sector [63].