科股宝VIP是北京商报联合钛媒体共同推出的新闻资讯产品，包括科股一线解读、大事件必读、并购一 线、精华研报等数个栏目，围绕新闻及政策动态，提供媒体视角的解读。 以下是近期复盘及栏目发稿回顾： 1、 AI链、海内外算力链、国产半导体企业IPO迎来集中进展、部分存储产品涨价 （1）6月5日，栏目组发文《市值超越微软！英伟达3.46万亿美元"封王"，AI产业链需求持续爆发丨科 股一线拆解》文章指出： 英伟达市值再度超越微软，重夺全球市值最高上市公司头衔。截至6月4日收盘，英伟达股价报 141.92美元，过去一个月累计涨幅达25.59%，总市值达3.46万亿美元。 这场AI盛宴还惠及更多科技公司。高速连接以及高速串行链路通信解决方案提供商CRDO Credo Technology也于6月3日盘中暴涨30%。 伴随着AI算力需求的结构性爆发，光模块/CPO、液冷服务器、AEC、PCB/CCL等业绩兑现确定性 较强的AI上游算力产业链可能率先响应。 AI中下游领域还需等待边际改善。偏硬件的上游算力和下游端侧AI环节已经率先兑现业绩，中游 软件、下游AI应用等环节则尚未迎来困境反转。 国内来看，未来两周有两个重要议程：6月 ...

文|汽车像素(ID:autopix)，作者 |曹琳，编辑|冒诗阳 01.东风多研发部门拆入奕派 ▍奕派科技总经理汪俊君 6 月 26 日，东风汽车集团首次披露了将成立 "东风奕派汽车科技有限公司"（下称奕派科技）的消息。 奕派科技代表着东风汽车集团下属自主乘用车品牌向新能源转型的希望，风神、奕派、纳米三个嫡系品牌都被纳入新公司。围绕着这家新公司的运作、职 能分配，"奕派独立" 是这周末里备受东风职工关注的话题。 调整不止涉及东风乘用车一个板块，还包括研发、制造、销售在内的多个部门和公司主体，其中的核心项是研发。 汽车像素独家了解到，东风汽车研发总院（下称研发总院）的副院长程军将加入奕派科技，担任新公司的总工程师。与乘用车相关的大部分研发团队，会 从研发总院中分拆出来，并入新成立的奕派科技，具体包括乘用车开发中心、海外车型平台项目团队两个部门。 除了上述两部门的完整编制，东风汽车还会从研发总院的其他中心里抽调技术精英加入奕派科技，以快速补充新公司所需。乘用车开发中心是研发总院最 大的部门之一，一位接近东风汽车的负责人告诉我们，最终研发总院三分之一到近半数人员可能都会被划至奕派科技。 母公司东风汽车希望赋予奕派科 ...

Core Viewpoint - Bilibili (B站) serves as a significant cultural barometer within the rapidly evolving online subculture, showcasing diverse narratives and creators that reflect the complexities of contemporary youth culture [1][5]. Group 1: Key Creators and Their Narratives - "峰哥亡命天涯" (Feng Ge) and "户晨风" (Hu Chenfeng) represent two contrasting yet influential figures in Bilibili, embodying different aspects of youth culture [2][10]. - Feng Ge is characterized by an "abstract" and "madness" style, deconstructing mainstream values and presenting a philosophy of failure and authenticity through his content [6][7]. - Hu Chenfeng, in contrast, focuses on social justice and the struggles of the lower class, using a grounded narrative style that resonates with viewers seeking representation and moral engagement [10][12]. Group 2: Cultural Significance and Youth Engagement - The narratives of Feng Ge and Hu Chenfeng create a sociological text that reflects the tensions within contemporary youth culture, balancing between idealism and cynicism [5][15]. - Feng Ge's approach offers a space for young people to embrace failure and reject societal pressures, while Hu Chenfeng provides a voice for social justice and moral responsibility [9][16]. - The coexistence of these two figures illustrates a complex cultural phenomenon where youth navigate their identities and anxieties in a digital landscape, seeking both humor and meaning [17][18].

一小时订单量超越宝马半年销量，属于小米汽车的高光时刻再现，中国智能汽车产业的历史正在被改 写。 2025年6月27日，小米集团-W（01810.HK）股价开盘飙升8%，突破61港元关口，再次刷新历史新高。 资本市场的热烈反应直指前一日发生的一场行业"地震"——小米首款SUV车型YU7上市仅3分钟大定订 单突破20万台，1小时飙升至28.9万台。 按照目前的订单数据，小米汽车今年再度超额完成年度销量目标的确定性极高。截止2025年5月21日， 小米SU7系列累计交付已超过25.8万台。 文 | 港股研究社 从"手机厂商"到"生态巨头"，万亿市值的小米正在重塑中国科技企业的成长范式。 花旗支招特斯拉：想和YU7拼？降价或送FSD、多给优惠 6月26日晚，北京亦庄的小米汽车工厂灯火通明。随着雷军公布小米YU7标准版25.35万元的起售价，订 单数据开始以几何级数飙升——3分钟突破20万台，1小时逼近29万台关口。 这一数字参考一年前小米SU7发布当晚4分钟大定破1万，跨越不可谓不大；而对比同行，更是超越了宝 马全系车型5个月在华销量（28.85万辆），完成特斯拉ModelY2024年在中国市场六成以上的销量 （48 ...

6月27日消息，今天上午举行的全球人工智能开发与应用大会上，腾讯云大数据基础产品中心总经理程 彬透露，腾讯云已经构建完善的"Data+AI"能力，今年下半年将发布数据智能体产品。 程彬表示，数据作为传统数据平台、AI大模型平台的共同"石油原料"，"Data+AI"不仅仅是数据的简单 叠加，更是深度融合与创新，实现从"让数据说话"到"与数据智能对话"的跃迁。 数据显示，2024年国内数据生产总量首次突破40ZB，达到41.06ZB，同比增长25%，增速较去年提高 2.56个百分点；人均数据生产量约为31.31太字节（TB），相当于1万多部高清电影，同比增长25.17%， 数据生产总量和人均产量实现同步跃升。 最新发布的《Data+AI下一代数智平台建设指南》报告显示，Gartner研究表明，非结构化数据占当今 组织数据的70%至90%。受生成式AI计划、多模态数据处理需求的爆炸式增长以及合规性压力的推 动，企业对非结构化数据管理的需求急剧增长。 因此，非结构化数据处理支出在数据管理总支出中所占的份额将越来越大。 随着AI大模型对于数据的需求增加，传统数据平台在应对生成式AI带来的新型数据需求时，正面临严 峻挑战 ...

Core Insights - Alibaba is undergoing significant organizational changes, merging Ele.me and Fliggy into its China e-commerce division to enhance its consumer platform strategy [1][2] - The company reported strong financial results for fiscal year 2025, with revenues reaching 996.347 billion yuan and a net profit increase of 77% to 125.976 billion yuan [1][2] - The shift towards a "big consumption platform" reflects a strategic upgrade aimed at optimizing business models and enhancing user experiences [1][9] Financial Performance - For fiscal year 2025, Alibaba's revenue was 996.347 billion yuan, with a net profit of 125.976 billion yuan, marking a 77% year-on-year increase [1][2] - Alibaba Cloud experienced double-digit revenue growth, driven by strong demand for AI-related products, which saw triple-digit year-on-year growth for seven consecutive quarters [1][2] - The international digital commerce group's revenue grew by 29% year-on-year, indicating robust performance in cross-border business [1][2] Business Adjustments - The merger of Ele.me and Fliggy into the China e-commerce division is part of a broader strategy to create a more integrated consumer experience [1][3] - The company has exited non-core assets like Hema and Intime, focusing on improving operational efficiency across its internet platforms [2][8] - A significant reduction in the number of partners from 26 to 17 reflects a shift towards a younger and more focused leadership team [2][9] Market Trends - The rise of instant retail is highlighted as a key trend, with Ele.me's integration into Alibaba's e-commerce strategy expected to enhance its competitive edge in this space [3][4] - The concept of "big consumption" emphasizes the importance of efficiently meeting existing consumer demands rather than solely pursuing new market opportunities [10][11] - The integration of online and offline channels is seen as essential for capturing consumer spending, with a focus on creating a seamless shopping experience [12][13]

Core Insights - The article discusses the current state and future of AI programming, highlighting skepticism about its capabilities and the challenges faced by developers in adopting AI tools [2][3][4] Group 1: AI Programming Capabilities - A recent benchmark test by a team of international algorithm competition winners revealed that top AI models like GPT-4o, DeepSeek R1, and Claude 3 had a 0% pass rate on high-difficulty programming problems when not allowed to use online information [2] - Developers express that while AI tools can enhance efficiency, they often require significant human oversight and cannot fully replace human programmers [4][8] - Many developers are still hesitant to trust AI-generated code, with a third of them not reviewing AI-generated code before deployment, raising concerns about security vulnerabilities [4][8] Group 2: Adoption Challenges - Companies face internal conflicts regarding the use of AI tools, with security departments often prohibiting their use while business units push for their adoption to improve performance [3][4] - The high cost of AI programming tools makes it difficult for companies to justify additional spending, especially when they are already at their IT budget limits [4][5] - Some companies have begun to develop their own AI tools to address specific needs and security concerns, as seen with ByteDance and Meituan [10][11] Group 3: Market Dynamics - Major companies like Goldman Sachs have invested significantly in AI tools like GitHub Copilot, spending millions annually, while also exploring competitive products [5][18] - The competitive landscape for AI programming tools is intensifying, with companies like Cursor and Windsurf emerging as significant players in the market [18][19] - Domestic AI programming tools are gaining traction, with improvements in model capabilities and a focus on data security and compliance, potentially narrowing the gap with international products [19]

Core Insights - Dazhong Dianping's "Must-Eat List" is becoming a significant indicator in the dining industry, showcasing 3,091 restaurants across 144 cities, with over half being first-time entrants [2][3] - The list reflects a growing trend of consumers actively seeking out local flavors and sharing their experiences, with 3.63 billion real reviews contributing to the platform's credibility [2][3][8] - The dining landscape is evolving, with a notable increase in user engagement and a shift towards exploring local culinary gems, as evidenced by the 78 billion active searches for food in the past year [2][3] Summary by Sections Must-Eat List Overview - The 2025 "Must-Eat List" was unveiled in Nanjing, featuring 3,091 restaurants, with Shanghai, Beijing, Chengdu, Chongqing, and Shenzhen leading in the number of entries [2] - Over 2,400 of the listed restaurants are lesser-known local spots, while more than 1,200 are long-established eateries, with over 40% being over ten years old [2] Consumer Behavior Trends - Consumers are increasingly willing to explore local neighborhoods for authentic dining experiences, as shown by a 50% year-on-year increase in reviews for restaurants over ten years old [3][4] - The concept of "taste travel" is gaining traction, with over 60% of users in listed cities being from outside those areas, indicating a growing demand for culinary exploration [5] Evaluation Process - The "Must-Eat List" is built on a rigorous evaluation process involving data engineers and product managers, ensuring high credibility and objectivity in the selection of restaurants [8][9] - In the past year, the platform saw a nearly 60% increase in real reviews, with 3.63 billion evaluations used to select the 3,091 restaurants from over 6 million merchants [8][9] Impact on Local Businesses - The list has significantly boosted visibility for local eateries, with new entrants experiencing a 50% increase in user traffic and overall transaction volume following their inclusion [11][12] - Many small businesses, previously unaware of the list, have benefited from the exposure, leading to increased customer flow and engagement [11][12] Industry Trends - The dining industry is witnessing a shift towards authenticity and quality, with many restaurants focusing on fresh ingredients and customer experience rather than chasing artificial ratings [12][13] - The "Must-Eat List" serves as a vital tool for restaurants to understand consumer preferences and market trends, promoting a return to core culinary values [12][13]

Core Insights - The SUV market for electric vehicles priced above 200,000 yuan is dominated by models like the Tesla Model Y, which outsells the Model 3 by nearly three times in 2024 [1] - Xiaomi's new model, the YU7, is positioned as a competitor to the Model Y and other domestic brands, following the success of the SU7 [1][2] - The YU7 has received significant initial orders, surpassing the SU7's performance, indicating strong market interest [4] Product Positioning - The YU7 is priced at 253,500 yuan, 279,900 yuan, and 329,900 yuan for its three versions, making it only 10,000 yuan less than the Model Y [1][16] - The YU7 features a more moderate design strategy compared to the performance-oriented SU7, focusing on practicality and comfort [6] - The YU7's dimensions are 4999mm in length, 1996mm in width, and 1608mm in height, classifying it as a mid-large SUV with a sporty design [6] Features and Specifications - The YU7 boasts upgraded interior features, including zero-gravity seats and a new panoramic display, enhancing user experience [8] - Performance specifications include a maximum power of 690 PS, a 0-100 km/h acceleration time of 3.23 seconds, and a maximum range of 835 km [10] - Advanced safety features include a robust structure with enhanced pillar strength and a comprehensive suite of smart driving technologies [14][12] Market Dynamics - The YU7's launch comes amid fierce competition in the SUV segment, particularly from domestic brands like Zeekr and NIO [16][24] - The YU7's success may not directly impact Tesla's Model Y but could significantly affect the sales of other domestic brands [20][21] - The YU7's initial order volume of over 122,000 units within three minutes of launch indicates strong market demand [4][25] Production Capacity Challenges - Xiaomi's production capacity is a critical factor, with the current output for the SU7 being less than 30,000 units per month, which may limit the YU7's delivery capabilities [25] - The second factory is expected to add 150,000 units of annual capacity, but achieving production levels comparable to the Model Y remains a challenge [25] - The ability to meet high demand while managing production constraints will be crucial for Xiaomi's market positioning [25]

Group 1: Cross-Border Payment Market - The global cross-border payment market is expanding, with multiple banks entering the trillion-yuan market following the launch of the Cross-Border Payment platform [2] - According to EIU, the global cross-border payment transaction volume is expected to reach $194.6 trillion in 2024, maintaining steady growth compared to 2023, and projected to grow at a CAGR of 6.4% to $320 trillion by 2032 [2] Group 2: Solid-State Battery Technology - EVTank predicts that global solid-state battery shipments will reach 614.1 GWh by 2030, with a market size exceeding 250 billion yuan [3] - The upgrade of battery technology is a core driver of expanding terminal demand, with solid-state batteries being identified as the next-generation battery technology due to their high energy density and safety [3] Group 3: Nvidia's Market Valuation - Loop Capital forecasts that Nvidia's market valuation could expand to $6 trillion, benefiting from a new wave of AI application growth, representing a 65% increase from its current valuation of $3.6 trillion [4] - The demand for computing power is spreading globally, with North American infrastructure companies frequently exceeding performance expectations [4] Group 4: AI in Education - The Tianli International-hosted AI model application results conference will showcase innovations in AI education technology, emphasizing its role in reshaping the educational ecosystem [5] - The "AI + Education" market is projected to reach 160 billion yuan by 2027 and nearly 180 billion yuan by 2030, driven by technological advancements and increased funding attention [5]