公告称，集团截至2025年6月30日止六个月的未经审核综合中期业绩预期收益减少及净亏损增加主要由 于：(1)为增强长期竞争力，集团加速从品牌代运营向产业链上游延伸，强化供应链自主可控能力及自 有商品矩阵建设，自有品牌的产品研发和市场拓展投入加大，导致经营成本阶段性上升，且短期内尚未 进入全面收益转化期，致使公司利润承压;及(2)公司计提财务资产预期信贷损失，导致亏损额增加。 豆盟科技(01917)发布公告，预期(i)截至2025年6月30日止六个月的收益将约人民币1800万元至人民币 1900万元，而去年同期的收益约为人民币2020万元;及(ii)截至2025年6月30日止六个月的净亏损将约人民 币1150万元至人民币1250万元，而去年同期的净亏损约为人民币410万元。 ...

Group 1 - On August 1, the Hong Kong stock market saw a net inflow of 12.207 billion HKD from northbound trading, with 5.575 billion HKD from the Shanghai Stock Connect and 6.632 billion HKD from the Shenzhen Stock Connect [2] - The most bought stocks included the Tracker Fund of Hong Kong (02800), Hang Seng China Enterprises (02828), and Xiaomi Group-W (01810), while the most sold stocks were Innovent Biologics (01801), Alibaba-W (09988), and SMIC (00981) [2] - The Tracker Fund of Hong Kong received a net inflow of 37.44 billion HKD, Hang Seng China Enterprises received 17.19 billion HKD, and Southern Hang Seng Technology (03033) received 8.55 billion HKD [5] Group 2 - Xiaomi Group-W (01810) saw a net inflow of 8.55 billion HKD, with the chairman announcing that July's car deliveries exceeded 30,000 units due to increased production capacity [6] - Tencent (00700), Meituan-W (03690), and Kuaishou-W (01024) received net inflows of 8.36 billion HKD, 7.66 billion HKD, and 2.02 billion HKD respectively, while Alibaba-W (09988) faced a net outflow of 2.02 billion HKD [6] - Innovent Biologics (02577) received a net inflow of 3.63 billion HKD after being listed as a partner by NVIDIA for its 800V direct current power architecture [7] Group 3 - Guotai Junan International (01788) experienced a net outflow of 48.99 million HKD amid the implementation of the stablecoin regulatory framework in Hong Kong [8] - Other stocks that saw net inflows included CSPC Pharmaceutical Group (01093) with 2.52 billion HKD and Giant Legend (06683) with 6.55 million HKD, while Innovent Biologics (01801) and SMIC (00981) faced net outflows of 4.02 billion HKD and 1.77 billion HKD respectively [8]

Group 1 - On August 1, 2025, InnoCare Pharma (02577), Tencent Holdings (00700), and Alibaba Group-W (09988) ranked as the top three companies by trading volume in the Southbound Stock Connect, with transaction amounts of 3.333 billion, 3.023 billion, and 2.736 billion respectively [1] - In the Southbound Stock Connect, Tencent Holdings (00700), Yingfu Fund (02800), and Southern Hang Seng Technology (03033) were the top three companies by trading volume, with transaction amounts of 2.389 billion, 2.076 billion, and 1.854 billion respectively [1] Group 2 - The top ten active companies in the Southbound Stock Connect included InnoCare Pharma (02577) with a transaction amount of 3.333 billion and a net buying amount of +320 million, Tencent Holdings (00700) with 3.023 billion and +102 million, and Alibaba Group-W (09988) with 2.736 billion and -416 million [2] - In the Southbound Stock Connect, Tencent Holdings (00700) had a transaction amount of 2.389 billion and a net buying amount of +734 million, Yingfu Fund (02800) had 2.076 billion and +2069 million, and Southern Hang Seng Technology (03033) had 1.854 billion and +856 million [2]

Group 1: Core Insights - The AI arms race is escalating rapidly, with Wall Street surprisingly applauding massive capital expenditures by tech giants [1][2] - Major tech companies like Meta, Microsoft, Google, and Amazon are expected to spend nearly $400 billion on AI infrastructure this year, surpassing the EU's total defense spending last year [1][3] - These investments are projected to contribute up to 0.5 percentage points to US GDP growth this year and next [1] Group 2: Winners in the AI Race - Meta has seen its AI investments translate directly into increased advertising revenue, leading to a stock price surge and a market cap increase of approximately $200 billion [4] - Microsoft reported a record capital expenditure of $30 billion for the quarter, with Azure's annual sales exceeding $75 billion, showcasing the returns from its AI investments [5] - Google's significant capital expenditure increase to $85 billion has not negatively impacted its revenue, with AI features driving a 10% increase in user queries [5] Group 3: Challenges Faced by Some Giants - Amazon's AWS cloud business is experiencing slower growth compared to competitors, raising doubts about its AI strategy despite a capital expenditure of around $118 billion [8] - Apple is perceived as lagging in AI investments, with internal challenges affecting its innovation capabilities, leading analysts to suggest acquisitions as a potential solution [6][7]

Core Points - WeChat has quietly reduced its withdrawal fees, with the minimum service fee now set at 0.01 yuan [1][6] - The previous policy allowed users a lifetime free withdrawal limit of 1000 yuan, with fees of 0.1% applied to amounts exceeding this limit, and a minimum fee of 0.1 yuan per transaction [4][6] - The adjustment significantly lowers costs for small withdrawals, benefiting more users and enhancing the flexibility of small fund usage [6] Summary by Sections Fee Structure Changes - As of July 25, WeChat updated its fee structure, setting the minimum service fee for withdrawals to 0.01 yuan [1] - Previously, users faced a minimum fee of 0.1 yuan for withdrawals exceeding the 1000 yuan free limit [4] Comparison with Competitors - Alipay also implemented withdrawal fees starting October 12, 2016, with a higher free limit of 20,000 yuan for users with multiple accounts under the same ID [4] - Both platforms charge a fee of 0.1% for amounts exceeding their respective free limits, with a minimum fee of 0.1 yuan per transaction [4] Rationale Behind Fees - WeChat's official response indicated that the fees are intended to balance transaction costs incurred by banks, which charge WeChat for various payment functionalities [4]

Core Viewpoint - The article discusses the investment journey of Masayoshi Son, highlighting his significant financial losses and remarkable recoveries, emphasizing the dual nature of his investment philosophy: taking bold risks and the potential for both great gains and substantial losses [5][6][34]. Group 1: Investment Philosophy - Masayoshi Son's investment strategy is characterized by a willingness to take significant risks, often leading to substantial financial losses, as seen in his history of both winning and losing large sums [6][26]. - The article contrasts Son's approach with the more conservative investment principles of Warren Buffett, suggesting that Son thrives in volatile environments where he can capitalize on opportunities others might avoid [7][50]. Group 2: Key Milestones - Son's career is marked by five pivotal moments: the rise of personal computers, the internet bubble, the rise of China, the global financial crisis, and the emergence of artificial intelligence [15][22]. - His early investments, such as in Yahoo, yielded significant returns, showcasing his ability to identify and capitalize on emerging trends [19][29]. Group 3: Recent Developments - In 2023, Son's investment in ARM, which went public, marked a significant recovery for SoftBank, with ARM's market value exceeding $150 billion, reflecting Son's ongoing influence in the tech sector [40][48]. - Despite past failures, such as the WeWork debacle, Son continues to pursue ambitious projects, including a $500 billion investment plan aimed at advancing AI and technology in Japan [9][44]. Group 4: Challenges and Future Outlook - The article highlights the challenges Son faces in the current investment landscape, particularly in the AI sector, where competition from major tech companies has intensified [41][48]. - Son's vision for Japan's role in the AI race is hampered by a lack of talent, prompting him to seek partnerships to bolster Japan's capabilities in this critical area [44][45].

港股通(深)活跃成交股 智通财经APP获悉，8月1日港股市场，北水成交净买入122.07亿港元，其中港股通(沪)成交净买入55.75 亿港元，港股通(深)成交净买入66.32亿港元。 北水净买入最多的个股是盈富基金(02800)、恒生中国企业(02828)、小米集团-W(01810)。北水净卖出最 多的个股是信达生物(01801)、阿里巴巴-W(09988)、中芯国际(00981)。 | 股票名称 | 买入额 | 卖出额 | 买卖总额 | | --- | --- | --- | --- | | | | | 净流入 | | 英诺赛科 | 18.26亿 | 15.06 乙 | 33.33亿 | | HK 02577 | | | +3.20 亿 | | 腾讯控股 | 15.62 乙 | 14.60亿 | 30.23亿 | | HK 00700 | | | +1.02 乙 | | 阿里巴巴-W | 11.60 乙 | 15.76 Z | 27.36亿 | | HK 09988 | | | -4.16 Z | | 国泰君安 ... | 11.71亿 | 12.20亿 | 23.91亿 | | HK 01788 | | ...

对此，Google回应第一财经记者问询时表示，该截图不是来自Google。 网传"谷歌中国公告"的截图，称自2025年9月1日起，Google获准全面恢复在中国大陆地区的服务。 ...

Core Viewpoint - The current market is not characterized by "mean reversion," with AI significantly impacting tech giants, leading to better-than-expected performance and increased capital expenditures [1][2]. Group 1: Company Performance - Meta Platforms reported impressive earnings, with advertising revenue growth accelerating by 2 percentage points to 22% year-over-year, driven by AI improvements [3][7]. - Microsoft's recent earnings report was outstanding, maintaining stable gross and operating margins despite significant capital expenditure increases and AI-driven revenue growth [4]. - AI investments are yielding clear returns for Meta, enhancing the efficiency and profitability of its advertising systems [5][7]. Group 2: Capital Expenditure Trends - Google raised its 2026 capital expenditure forecast by approximately $18 billion to $102 billion, while Meta and Microsoft also significantly increased their capital expenditures [8]. - Meta's 2026 capital expenditure was adjusted upward by about $25 billion to $100 billion, and Microsoft's was raised to approximately $116 billion for 2026 [8]. Group 3: Software Industry Insights - The software industry is showing mixed results, with strong performances from Microsoft and ServiceNow, but disappointing results from Check Point and Confluent, indicating a challenging market environment [9]. - The software sector is described as a "stock-picking market," reflecting the ongoing volatility and lack of a clear upward trend [9]. Group 4: Cloud Services Growth - Public cloud services remain a major growth theme, with Microsoft's Azure revenue growth accelerating by 4 percentage points to 39% year-over-year, amid ongoing capacity shortages [10]. - Microsoft Fabric has seen a 55% year-over-year growth, with over 25,000 users, indicating strong demand for cloud services [10]. Group 5: AI and Internet Sector - The internet sector's sentiment towards AI is generally moderate, with companies leveraging AI to enhance user experiences, as seen with Meta and Booking Holdings [11]. - The narrative around AI assistants is gaining traction, showcasing the potential for personalized services in the internet space [11]. Group 6: Market Dynamics - The market is increasingly characterized by a "stronger gets stronger" dynamic, where favored AI-related large-cap stocks continue to attract capital and outperform [12]. - Companies that have lagged behind, even with lowered expectations or valuations, continue to struggle post-earnings announcements [12].

Core Viewpoint - The article discusses the innovative recommendation algorithm architecture, RankMixer, developed by ByteDance, which significantly enhances the efficiency and effectiveness of video recommendations on platforms like Douyin while maintaining low inference costs [2][40]. Group 1: RankMixer Model Overview - RankMixer represents a new recommendation model architecture that increases the parameter scale from tens of millions (16M) to billions (1B), enhancing model performance without increasing inference latency [4][26]. - The model design focuses on aligning with GPU hardware characteristics, allowing for efficient computation through large matrix multiplications, thus overcoming memory bottlenecks [9][41]. - RankMixer incorporates innovative features such as TokenMixing and Per-Token SparseMoE, which improve the model's ability to capture diverse feature interactions and enhance parameter efficiency [12][24]. Group 2: Performance Metrics and Improvements - In the Douyin recommendation scenario, the RankMixer-1B model has shown a cumulative increase of over 0.3% in user active days and more than 1% in average daily usage time, indicating improved user engagement [4][35]. - The model's efficiency is highlighted by a 70-fold increase in parameters while keeping the inference cost stable, achieved through various optimization techniques [26][30]. - Offline metrics show that RankMixer-1B outperforms traditional DNN models, with an AUC increase of over 0.9% and UAUC improvement exceeding 1% [32]. Group 3: Technical Innovations - RankMixer employs Automatic Feature Tokenization to align input features into a uniform token sequence, facilitating parallel processing and maximizing hardware utilization [15][16]. - The TokenMixing module allows for efficient information exchange between tokens, enhancing the model's ability to leverage global information for better recommendations [19][20]. - The Per-Token SparseMoE architecture enables differentiated modeling of semantic subspaces, significantly increasing parameter capacity while reducing computational overhead [21][24]. Group 4: Future Implications - The successful implementation of RankMixer across various ByteDance applications demonstrates its potential as a universal ranking model architecture [39]. - The exploration of RankMixer validates the importance of co-designing algorithms with infrastructure to optimize machine learning performance and resource utilization [43][44].