Group 1 - DeepSeek's emergence accelerates AI application proliferation and boosts chip companies' technological iteration, injecting growth momentum into the Sci-Tech Innovation Board [1] - As of May 30, 2025, the Sci-Tech Innovation Board has 587 listed companies with a total market capitalization of 6.64 trillion yuan, establishing itself as a core area for cutting-edge technology in China [1] - The launch of the TaiKang Sci-Tech Innovation Index Enhanced Fund on June 9 aims to help investors seize opportunities in the "Sci-Tech + Broad-based" index [1] Group 2 - The Sci-Tech Innovation Index covers 570 constituent stocks, representing over 97% of the total market capitalization of the Sci-Tech Innovation Board, providing a comprehensive view of China's technology innovation industry [2] - The index encompasses all 16 primary industries and 44 secondary industries, achieving a coverage rate of 96%, surpassing existing indices like Sci-Tech 50, 100, and 200 [5] Group 3 - The average and median market capitalization of the Sci-Tech Innovation Index constituents are 12.3 billion yuan and 5.3 billion yuan, respectively, aligning closely with the overall market [9] - The median R&D expense as a percentage of revenue for the index constituents is 12.5%, significantly higher than other broad-based indices, indicating strong innovation vitality [9] Group 4 - The expected profit growth for the Sci-Tech Innovation Index is projected to reach 84.95% in 2024, with an annualized growth rate of 39.26% from 2024 to 2025, outpacing the Shanghai and Shenzhen indices [12] - Since its base date of December 31, 2019, the Sci-Tech price index has achieved a cumulative return of 13.7%, outperforming the Sci-Tech 50, 100, and the CSI 300 indices [15] Group 5 - Current market conditions favor the allocation to the Sci-Tech Innovation Index, supported by strong policy backing and an expected global liquidity increase due to potential interest rate cuts by the Federal Reserve [18] - The core industries within the Sci-Tech Innovation Board are experiencing positive trends, with sectors like semiconductors and AI applications showing signs of recovery and growth [18] Group 6 - The rapid inclusion mechanism for new stocks on the Sci-Tech Innovation Board enhances the potential for excess returns, providing more alpha opportunities for actively managed index funds [19] - The Sci-Tech Innovation Index's diversified characteristics reduce investment risks associated with single market cap fluctuations, making it suitable for various market environments [15] Group 7 - The TaiKang Sci-Tech Innovation Index Enhanced Fund aims to leverage a dual strategy of index Beta and quantitative Alpha enhancement to achieve stable growth in returns [21] - The fund is managed by an experienced fund manager, Yuan Shuai, who has demonstrated strong performance in managing various funds across A-share and Hong Kong markets [22] Group 8 - The Sci-Tech Innovation Board is positioned as a key area for technological breakthroughs and an important direction for asset allocation, with the TaiKang Sci-Tech Innovation Index Enhanced Fund being launched to facilitate investment in future technologies [23]

Market Overview - On June 10, the three major stock indices opened mixed, with the Shanghai Composite Index rising by 0.07%, the Shenzhen Component Index falling by 0.02%, and the ChiNext Index declining by 0.16% [1] Guest Opinions - Guotai Junan Securities' investment advisor Wang Chen believes that the overall market risk is controllable, and a short-term upward trend is expected. A key upward breakthrough may require the activation of heavyweight sectors [2] - Dongfang Securities' investment advisor Ying Fang suggests that investors focus on specific sectors that have undergone sufficient adjustments, while also considering taking profits in sectors with long-term potential [2] - Jinyuan Securities' chief investment advisor Xu Chuanbao holds an optimistic view of the market, recommending that investors seize structural opportunities and select quality stocks for medium to long-term investment [2] Brokerage Insights - CITIC Securities reports a slight decline in export growth, with May's export growth rate dropping by 3.3 percentage points to 4.8%, slightly below market expectations. The report highlights strong export growth to non-US markets and the significant contribution of the semiconductor and automotive industries to exports [5][6] - Huatai Securities indicates that AI applications are expected to see rapid growth in areas where data processes are standardized and ROI is quantifiable. The commercialization of AI is accelerating, particularly in ToC scenarios like gaming and e-commerce, while ToB applications in marketing and sales are moving towards large-scale implementation [7]

Group 1 - The core point of the news is that the computing power concept stocks are leading the market, with Haiguang Information planning a 116 billion yuan stock swap to absorb and merge with Zhongke Shuguang, resulting in significant stock price increases for both companies [1] - The A-share market opened with mixed results, with the Shanghai Composite Index up by 0.07% and the Shenzhen Component Index down by 0.02% [1][2] - Haiguang Information's stock surged over 8%, while Zhongke Shuguang and Zhuguang Shucai reached their daily limit up [1] Group 2 - The Hang Seng Index opened up by 0.21%, with notable gains in technology stocks, particularly Kuaishou, which rose nearly 3% [3] - The People's Bank of China conducted a 198.6 billion yuan reverse repurchase operation with a rate of 1.40% [4] - The central bank adjusted the RMB to USD midpoint rate up by 15 basis points to 7.1840 [5]

Core Insights - The report emphasizes that the pessimistic expectations surrounding computing infrastructure are reversing, with leading companies like Nvidia and Broadcom showing strong market performance, indicating a sustained high prosperity trend in computing infrastructure [3][4][5] - The AI industry chain is expected to have a certain rebound potential under the influence of overseas mapping and market style rotation, making it an optimal time for investment [2][4][6] Computing Infrastructure - Recent performance in the A-share TMT sector has been notable, with telecommunications, computers, electronics, and media sectors rising by 2.42%, 2.42%, 2.23%, and 1.34% respectively, while computing infrastructure components like optical modules, copper connections, and circuit boards saw increases of 5.18%, 4.45%, and 3.88% [4] - The launch of new AI products, such as AR glasses, is accelerating the integration of AI applications from concept to execution, with major tech companies like Microsoft and Google heavily investing in this area [6] AI Industry Chain - The report highlights that the AI application landscape is transitioning towards deep integration and task execution, with significant developments expected at the upcoming ByteDance Volcano Engine 2025 conference [6] - The computing infrastructure sector, particularly components directly benefiting from the high prosperity of overseas computing leaders like Nvidia, is identified as a key area for investment focus [6]

[Table_Title] 研究报告 Research Report 8 Jun 2025 香港策略 Hong Kong Strategy 宽幅震荡升级在即，继续回避微盘股和新消费风险 Broad Consolidation About to Intensify; Remain Cautious on Micro-Caps and New Consumer Names 周林泓 Amber Zhou 李加惠 Jiahui Li, CFA amber.lh.zhou@htisec.com jh.li@htisec.com [Table_yemei1] 观点聚焦 Investment Focus [Table_summary] (Please see APPENDIX 1 for English summary) 上周我们认为当前市场处于宽幅震荡中，回调仍未结束，尤其需要规避短期过热的A股小微盘股及港股新消费板 块。本周一，受中美贸易摩擦影响，恒生指数早盘一度跌近2.7%，恒生科技最多跌3.1%，两者从5月高点分别累计 回撤5.2%、8.9%。这波下跌充分释放了市场对中美贸易不确定性的担忧，因此市场进入宽幅震荡以来 ...

Core Insights - The article emphasizes the continuous improvement of AI applications driven by technological advancements, highlighting that certain sectors have achieved user growth, retention, and monetization, warranting sustained attention on key sectors and leading products [1][2]. Sector Comparison - The comprehensive assistant sector shows broad coverage, with deeper penetration in social, education, and office applications. Mature products attract user traffic, with AI assistants and image editing leading the way. Products like DeepSeek demonstrate breakthrough growth due to excellent performance [2]. - In terms of user stickiness, social interaction products have developed robust features that foster user engagement. The education sector exhibits strong payment habits, with customers willing to pay for high-quality products. Content production tools can command high prices due to significant computational demands and diverse needs [2]. Key Product Analysis - The comprehensive assistant sector is witnessing transformative changes, with products like Doubao introducing "video call" features, indicating a shift in AI interaction methods. Tencent's Yuanbao showcases the positive impact of its ecosystem on product capabilities and user acquisition [3]. - In the AI video domain, products such as Keli, Jimeng, and Hailuo have demonstrated excellent performance with stable user demand. Google's recent launch of VEO3 marks a new phase in "audio-visual synchronized generation," potentially guiding the next stage of AI video development [3]. - The popularity of AI Agents, particularly with Manus, has sparked interest, with domestic players like Kunlun Wanwei's Tiangong Super Intelligent Agent achieving impressive test scores. This suggests more opportunities in complex task completion [3]. - AI companionship products exhibit strong user stickiness, although monetization through UGC models remains a challenge. Future attention may be directed towards premium products like "EVE," which could offer new insights for the industry [3].

Investment Rating - The industry investment rating is "Overweight" [1][2] Core Insights - The development of AI applications is showing promising trends, with a focus on long-term advantages in specific sectors and quality products. The report maintains an "Overweight" rating for the media culture industry, suggesting attention to companies with excellent product layouts and potential [2][3]. - The evolution of AI applications has transitioned from "pure text" to "static images/sounds" and now to "dynamic videos," indicating a richening of application tracks. The technology is evolving from "single modality/single task" to "multi-modal comprehensive/complex tasks" [3][4]. Summary by Sections Application Trends - AI applications are experiencing a shift towards multi-modal capabilities, enhancing user engagement and interaction [3][4]. - The report highlights that mature products are gathering user traffic, with AI assistants and image editing tools leading the way. Products like DeepSeek are noted for their breakthrough growth due to excellent performance [4]. Key Product Analysis - The report emphasizes the importance of leading products in shaping industry trends. For instance, Doubao has introduced a "video call" feature, indicating a transformation in AI interaction methods. Other notable products include Keling, Jimeng, and Hailuo, which have established stable user demand [5][29]. - The AI agent Manus has gained attention for its ability to perform complex tasks independently, showcasing the potential of AI agents in various applications [50]. Market Performance - The report provides data on the monthly active users (MAU) of leading AI applications, with ChatGPT leading globally at 546.15 million MAU, followed by Quark and Doubao with 149.1 million and 107.28 million MAU respectively [19][31]. - The AI video sector has shown stable demand, with products like Keling and Hailuo achieving significant daily traffic and user engagement metrics [40][41]. Competitive Landscape - The report identifies key players in the AI assistant space, such as Doubao and Tencent's Yuanbao, which have integrated into their respective ecosystems, enhancing user experience and functionality [32][35]. - The AI companionship sector is highlighted for its mature products, with Character AI and Talkie AI maintaining substantial user engagement [57]. Future Outlook - The report suggests that the AI application landscape is evolving rapidly, with a focus on enhancing user interaction through innovative features and multi-modal capabilities. The introduction of products like EVE, which aims to provide a more immersive AI companionship experience, is noted as a potential game-changer [62][68].

Core Insights - The report from Guotai Junan highlights the continuous improvement of AI applications driven by technological advancements, with certain sectors already achieving user growth, retention, and monetization, suggesting a focus on key sectors and leading products [1][2] Group 1: AI Application Development - The evolution of AI applications has transitioned from "pure text" to "static images/sounds" and now to "dynamic videos," reflecting a richening of application sectors [2] - AI technology is evolving from "single modality/single task" to "multi-modal comprehensive/complex tasks," indicating a broader scope for AI applications [2] Group 2: Sector Comparison - Comprehensive assistants have a wide coverage, with deeper penetration in social, educational, and office-related uses; mature products attract user traffic, while products like DeepSeek show potential for breakthrough growth [3] - Social interaction products exhibit strong user engagement due to well-developed functionalities, leading to increased user stickiness [3] - The education sector shows a strong willingness to pay for quality products, with content production tools able to command high prices due to high computational demands [3] Group 3: Key Product Analysis - Attention is drawn to leading products shaping industry trends; for instance, Doubao has introduced a "video call" feature, signaling a shift in AI interaction methods [4] - Domestic AI video products like Keling, Yimeng, and Hailuo have demonstrated strong user demand, while Google's VEO3 has initiated a new phase in "audio-visual synchronized generation" [4] - AIAgent, following the success of Manus, is gaining attention, with Kunlun Wanwei's Tiangong Super Intelligent Agent achieving high test scores, indicating potential for more opportunities in complex task completion [4] - AI companionship products show notable user stickiness, although monetization under UGC models remains a challenge; products like "EVE" may offer new insights for the industry [4]

Group 1 - The core viewpoint of the article highlights that by March 2025, the monthly active users of mobile AI applications in China will exceed 647 million, indicating that over half of Chinese internet users have entered the "AI-native application" era, with educational AI emerging strongly in this technological wave [1] - The success of Xiaoyuan AI, designed specifically for primary and secondary school students, is attributed to a deep understanding of the essence of education, as evidenced by its top position in user growth within the education AI sector and its entry into the top 10 newly downloaded AI applications across the internet [1][2] - Xiaoyuan AI is built on a robust dataset accumulated by Yuanfudao Group over more than a decade, which includes over 5 million hours of teaching videos, more than 2 billion question bank entries, over 10,000 knowledge points, and 1 million exam papers, along with 30 billion learning data points from 500 million global users, providing a solid foundation for high-precision learning analysis and personalized learning path planning [2][3] Group 2 - The core layer of Xiaoyuan AI integrates "educational genes" with "technological evolution" through the collaboration of self-developed Yuanli model and the top reasoning model Deepseek-R1, creating a unique model matrix that accurately identifies the root causes of student errors and transforms hard data into guided teaching that aligns with cognitive principles [3][5] - Xiaoyuan AI APP offers a comprehensive learning solution for primary and secondary students, covering over 100 key learning scenarios such as homework checking, learning diagnostics, error analysis, and 1v1 explanations, with homework correction serving as a core scenario and the 1v1 personalized explanation feature achieving significant penetration in the past month, becoming a trusted learning assistant for parents and students [5]

Investment Rating - The report downgrades the investment rating to "Outperform" [2][9] Core Views - The company is experiencing performance pressure due to strategic adjustments, with revenue declining as it focuses on high-value clients and reduces low-margin businesses [7] - The gross margin is expected to recover to normal levels in Q1 2025 after a temporary reduction in 2024 [7] - The company has launched multiple AI projects and partnerships, indicating a focus on innovation and market adaptation [7] Financial Data Summary - Total revenue for 2024 is projected at 398 million, with a significant decline of 36.7% year-on-year, and a net loss of 696 million [6] - For Q1 2025, revenue is expected to be 37 million, down 31.6% year-on-year, with a net loss of 124 million [7] - The company anticipates a gradual recovery in revenue, projecting 544 million for 2025 and 742 million for 2026, with a net loss of 455 million and 360 million respectively [6][7] Business Strategy and Adjustments - The company is strategically adjusting its business model, reducing its focus on low-margin sectors while enhancing its AI capabilities [7][8] - The proportion of revenue from smart governance has decreased from 54% to 36%, while the share from innovative AI and other sectors has increased from 10% to 37% [7] - The report highlights the successful launch of various AI models and projects, including partnerships with Huawei for integrated solutions [7][9] Market Comparison - The report compares the company with peers in the AI sector, noting its broad range of services and technological capabilities [8][13] - The average price-to-sales ratio for comparable companies is noted, with the company being valued at a PS of 14x, indicating potential growth [9][10]