Core Viewpoint - The stock of Jiantao Laminated Board (01888) has increased by over 6%, reaching HKD 12.1 with a trading volume of HKD 171 million, driven by positive developments in the AI and high-performance computing sectors [1] Group 1: Company Developments - Oracle recently released its FY26Q1 financial report, showing rapid growth in remaining performance obligations and other financial metrics [1] - Nvidia launched the new generation Rubin CPX chip system, enhancing AI coding and video processing capabilities [1] Group 2: Industry Trends - Southwest Securities noted that the rapid development of generative AI models and their accelerated application is leading to continuous iterations in AI computing infrastructure, which is expected to accelerate construction [1] - CICC indicated that the explosive growth in global computing power demand, driven by the rapid advancement of AI large models, is propelling the high-performance server market's rapid expansion, thereby increasing demand for upstream PCB and CCL substrate markets [1] - Open Source Securities mentioned that the price of copper-clad laminates has already increased in the second half of the year, with strong PCB demand likely supporting price rigidity, and the company's performance in the second half is expected to benefit [1] - The acceleration of high-end CCL and material capacity layout in 2026 is anticipated, with AI options expected to enter a realization phase, driving the valuation center upward, and actual financial contributions are likely to be concentrated in 2027 [1]

Core Viewpoint - The company reported a decline in revenue and net profit for the first half of 2025, primarily due to various factors impacting performance in the short term [1][2]. Financial Performance - In 2025 H1, the company achieved revenue of 3.72 billion yuan, a year-on-year decrease of 5.1% [1][2]. - The net profit attributable to shareholders was 120 million yuan, down 57.0% year-on-year [1][2]. - EBITDA for 2025 H1 was 720 million yuan, reflecting a decline of 1.8% year-on-year [2]. IDC Business Insights - IDC revenue for 2025 H1 was 1.07 billion yuan, a decrease of 2.1% year-on-year, with a gross margin of 32.7%, down 1.8 percentage points [3]. - The company accelerated the deployment of data centers, with 16,000 new cabinets installed, bringing the total to over 72,000 cabinets and an operational scale of approximately 320 MW [3]. - Specific projects include the completion of the first phase in Tianjin and ongoing construction in Inner Mongolia and the Yangtze River Delta region [3][4]. Cloud Computing Business - Cloud computing revenue for 2025 H1 was 2.57 billion yuan, down 7.7% year-on-year, with a gross margin of 7.8%, a decrease of 1.7 percentage points [4]. - The decline was attributed to client business adjustments and proactive optimization of the customer structure [4]. - The company is leveraging AI technology and has signed a strategic cooperation agreement to enhance its computing infrastructure for generative AI applications [4]. Investment Outlook - The company is expected to face short-term performance pressure due to increased depreciation and the disposal of specific operational assets related to public cloud services, which will reduce cloud computing revenue by approximately 100 million yuan in 2025 [2][5]. - However, the IDC industry is anticipated to rebound, driven by significant demand from AI, suggesting potential long-term benefits for the company [5]. - Projected net profits for 2025-2027 are 328 million yuan, 393 million yuan, and 476 million yuan, with corresponding P/E ratios of 79x, 65x, and 54x [5].

Core Insights - Oracle's stock surged over 40%, marking its largest single-day increase in 33 years, despite slightly missing revenue and earnings expectations in its Q1 FY2026 report [2][4][5] Financial Performance - Oracle reported Q1 FY2026 revenue of $14.93 billion, a 12% year-over-year increase, but below the consensus estimate of $15.04 billion [5] - Adjusted EPS was $1.47, slightly lower than the forecast of $1.48 [5] Key Metrics - Remaining Performance Obligations (RPO) soared 359% year-over-year to $455 billion, indicating a strong future revenue stream from signed contracts [7][8] - RPO is nearly 30 times Oracle's revenue over the past 12 months, highlighting a significant backlog of future cash flow [7] Cloud Services Growth - Cloud service revenue increased by 28% to $7.2 billion, with Infrastructure as a Service (IaaS) growing 55% year-over-year to $3.3 billion [10] - Oracle's OCI business is projected to grow by 77% this fiscal year, reaching $18 billion, surpassing previous estimates [10] AI Infrastructure Positioning - Oracle is positioned as a key player in the AI infrastructure race, capitalizing on the high demand for computing power driven by generative AI models [12] - The company has secured significant contracts with major players like OpenAI and Meta for AI hardware and infrastructure [14] Strategic Investments - Oracle plans to invest $35 billion in capital expenditures for FY2026, significantly higher than the previous year's $21 billion, to expand its data center capacity [17] - This investment is aimed at supporting the growing demand for AI infrastructure and is seen as a critical component of Oracle's future growth strategy [17] Competitive Advantages - Oracle's extensive database customer base, including 97% of Fortune 100 companies, provides a strong foundation for its cloud services [19] - The company's Autonomous Database offers automation and integration capabilities that are attractive to enterprises transitioning to AI [19] Market Impact - Oracle's stock surge has positively influenced related stocks like Nvidia and AMD, contributing to record highs in the S&P 500 and Nasdaq indices [22] - Analysts have responded by upgrading Oracle's rating to "Buy," with a target price of $368, recognizing its transformation into an AI powerhouse [22]

Core Insights - The global generative AI application market is experiencing significant growth, with nearly 1.7 billion downloads and in-app purchase (IAP) revenue reaching $1.9 billion in the first half of 2025, marking a 67% increase in downloads and a doubling of revenue compared to the previous half-year [12][26][19]. Market Overview - The Asia region is identified as the growth engine, with downloads increasing by 80% from the second half of 2024 to the first half of 2025, significantly outpacing Europe (51%) and North America (39%) [22][26]. - North America, despite a decrease in download share to 11%, leads in IAP revenue with $762 million, reflecting a 74% quarter-over-quarter growth [26][19]. User Engagement - Users spent over 15.6 billion hours on generative AI applications in the first half of 2025, averaging 86 million hours per day, with a total of 426 billion interactions, equating to approximately 50 interactions per person globally [19][12]. Application Trends - The report highlights that AI assistants are driving the majority of downloads, with 85% of downloads in the AI assistant and content generator categories coming from AI assistants like ChatGPT, Google Gemini, and DeepSeek [29][33]. - DeepSeek has surpassed ChatGPT in download numbers within its first six months of release, indicating a competitive landscape among AI applications [33][12]. Advertising and Revenue Growth - In the second quarter of 2025, generative AI advertising spending in the U.S. exceeded $200 million, a 121% year-over-year increase, with social media accounting for nearly 70% of this expenditure [12][19]. - The report indicates a strong monetization potential for generative AI applications, with significant IAP revenue growth observed across various regions, particularly in Latin America (147%), Asia (136%), and the Middle East (131%) [26][12]. User Demographics - The user base for generative AI applications remains predominantly young and male, with nearly 70% of ChatGPT users being male and 64% under the age of 35 [37][41]. - However, leading applications like ChatGPT and Microsoft Copilot have successfully diversified their user demographics, achieving a more balanced gender distribution [37][41].

Group 1 - Nvidia launched the Rubin CPX GPU designed for long-context inference, capable of processing millions of tokens at once, supporting software development and video generation tasks [1] - The Rubin CPX GPU will be part of the Vera Rubin NVL144 CPX platform, providing 8 exaflops of AI computing power, which is 7.5 times that of the GB300 NVL72 system [1] - The system features 100TB of high-speed memory and 1.7 PB/s memory bandwidth, expected to be available by the end of 2026, promising unprecedented performance and efficiency for long-context tasks [1] Group 2 - Claude introduced a significant update allowing direct creation and editing of Excel, Word, PPT, and PDF files, outputting usable file formats [2] - The system is equipped with a private computing environment capable of writing code to generate various documents, supporting advanced data analysis and file operations [2] - This functionality is available to Max, Team, and Enterprise users, with Pro users to gain access in a few weeks, allowing file uploads or demand descriptions for Claude to process [2] Group 3 - Tencent released the AI CLI tool CodeBuddy Code and opened public testing for CodeBuddy IDE, supporting unlimited use of the DeepSeek model [3] - The system is designed for professional engineers, enabling natural language-driven development and operations, supporting multi-agent collaboration and deep integration with Git/CI/CD [3] - AI programming is evolving towards L4-level AI software engineering, with CLI becoming the foundational infrastructure, showing a 40% reduction in coding time and an increase in AI code review contributions from 12% to 35% [3] Group 4 - Kuaishou launched the AIGC super employee Kwali, capable of generating complete short videos from a single sentence, automating the entire process from script to publication [4] - The system is driven by a multi-agent framework, including intent parsing, script generation, shot matching, and editing, integrated with a material library [4] - Kwali allows independent manipulation of video elements on a timeline, enabling rapid video production that previously required multiple teams [4] Group 5 - Fellou CE created a "seamless continuum experience," achieving continuous interaction, task decomposition, and memory continuity [5] - The system supports cross-application execution, multimodal conversion, and dynamic workflow orchestration, successfully applied in travel planning and content creation [6] - Fellou CE introduced core features like "deep search" and "visual report generation," enhancing user control and productivity [6] Group 6 - Tencent released the open-source text-to-image model "Hunyuan Image 2.1," supporting native 2K images and achieving industry-leading performance in semantic understanding and text generation [7] - The model can handle prompts of up to 1000 tokens, generating detailed scene descriptions and supporting various styles [7] - Hunyuan Image 2.1 utilizes a 32x compression VAE and dual text encoders to improve training stability, reducing inference steps from 100 to 8 [7] Group 7 - Google launched an AI system to assist researchers in writing "expert-level" scientific software, combining large language models with tree search algorithms [8] - The system acts as a "mutation" engine during the search process, integrating and reorganizing research ideas from scientific literature [8] - It has shown exceptional performance in genomics, geospatial analysis, and neuroscience, marking a shift from one-time code generation to quantifiable scientific goal-oriented software evolution [8] Group 8 - a16z partners discussed that agents are not universal but systems composed of multiple agents, each specializing in specific tasks, leading to microservices and domain specialization [9] - Experts are becoming the biggest beneficiaries of AI, achieving a tenfold productivity increase, changing the nature of work rather than just output [9] - Each platform transition alters the abstract layer of human-computer interaction, with AI revolutionizing workflows and creating numerous vertical entrepreneurial opportunities [9] Group 9 - Elon Musk revealed that the Optimus humanoid robot will have near-human dexterity, costing around $20,000, with challenges mainly in hardware design [10] - The Tesla AI5 chip is expected to achieve a 40-fold performance leap over AI4, with software upgrades enabling Tesla cars to exhibit "awareness" [10] - The third-generation Starship will have a payload capacity exceeding 100 tons, aiming for full reusability next year, with human self-sufficiency on Mars projected within 25 years [10]

Group 1 - Significant net purchases were made in Alibaba (HKD 4.511 billion), Bilibili (HKD 0.362 billion), and other companies, while major net sales were recorded for Pop Mart (HKD 1.711 billion) and others [1] - Southbound funds have continuously net purchased Alibaba for 14 days, totaling HKD 33.41389 billion, while Pop Mart and Xiaomi have seen net sales over the last three days [3] Group 2 - Alibaba launched a new product called "Gaode Street Ranking," which utilizes AI algorithms to quantify user behavior and create a credit system for offline services, enhancing visibility for lesser-known businesses [4] - Goldman Sachs reported that Bilibili's upcoming game "Three Kingdoms: Hundred Generals Card" is expected to achieve daily active users of one million, with a launch planned for the 2026 Spring Festival [4] Group 3 - The AI cloud market in China is projected to reach CNY 22.3 billion in the first half of 2025, with a significant growth forecast driven by generative AI [5] - Bank of America upgraded Kingsoft Cloud's rating to "Buy" with a target price of USD 20.4, while JPMorgan maintained a "Buy" rating and raised the target price to USD 19 [5] Group 4 - Xiaomi's marketing manager was dismissed for serious violations, including leaking confidential information, indicating potential internal governance issues [6]

Group 1: Core Insights - The year 2025 is widely regarded as the "Year of AI Agents," marking the mainstream adoption of agent-based AI technology [2][12] - Agent products are evolving from simple tools to complex, collaborative ecosystems that will significantly change how businesses automate complex tasks and make decisions [2][3] - There is a growing concern regarding user interest and renewal rates for previously popular agent products, attributed to issues such as exaggerated capabilities and unclear pricing structures [3][4] Group 2: Industry Trends - The integration of AI agents into enterprise software is expected to rise dramatically, with projections indicating that by 2028, approximately 33% of enterprise software applications will embed agent-based AI [12] - The financial sector is seen as a promising area for AI application, but it faces challenges such as budget constraints, talent shortages, and planning issues [8][9] - By 2026, it is anticipated that at least 30% of generative AI projects in the banking sector will be abandoned due to poor data quality and budget limitations [8][9] Group 3: Technological Developments - The emergence of integrated computing solutions, such as all-in-one generative AI stacks, is expected to enhance the capabilities of AI agents, allowing for faster iterations and improved decision-making [6][7] - The development of specialized models tailored to specific industries or tasks is anticipated to increase performance and reduce issues like hallucinations in AI outputs [7][11] - The financial industry is moving towards a new phase of AI application, transitioning from basic automation to more autonomous decision-making capabilities [11][12]

Investment Rating - The industry investment rating is "Overweight" with a maintained rating from the previous assessment [7]. Core Insights - The report highlights the significant advancements in AI computing capabilities with the introduction of the Rubin CPX chip by Nvidia, which is expected to provide 30 million trillion FP4 precision floating-point operations and enhance video processing [2]. - Oracle's financial performance shows strong growth, with a 12% year-over-year increase in revenue to $14.93 billion and a 55% increase in cloud infrastructure revenue [4]. - The report suggests that leading companies in the AI computing infrastructure sector are likely to experience substantial and reliable growth due to the rapid development of generative AI models and their applications [4]. Summary by Sections Nvidia Rubin CPX Chip - The Rubin CPX chip is set to launch by the end of 2026 and can be integrated into existing server designs or operate as a standalone computing device [2]. - It features a unique inference architecture that separates the computation process into context and generation stages, optimizing resource allocation and achieving three times the attention acceleration performance [2]. - The chip is expected to transform software development by enabling AI systems to manage entire codebases and understand complex software project structures [3]. Oracle Financial Performance - In FY26Q1, Oracle reported revenues of $14.93 billion, a 12% increase year-over-year, with a non-GAAP net profit of $4.3 billion, up 8% [4]. - The cloud infrastructure segment generated $3.3 billion in revenue, reflecting a 55% year-over-year growth, while cloud applications revenue reached $3.8 billion, an 11% increase [4]. - Oracle's remaining performance obligations surged by 359% to $455 billion, indicating strong future revenue potential [4]. Investment Opportunities - The report recommends focusing on companies in the AI computing sector, including AIDC firms like SenseTime and Yunsai Zhili, AI chip manufacturers like Cambricon and Haiguang Information, and AI server providers like Inspur and Huaqin Technology [4].

Core Viewpoint - Apple has introduced new self-developed chips, including the A19 mobile chip, N1 wireless chip, and C1X modem, which enhance the performance and efficiency of the new iPhone Air, making it the most energy-efficient iPhone to date [1][5][27]. Group 1: Chip Developments - The A19 Pro chip features a new 6-core CPU and a 5-core GPU, making it the fastest CPU among smartphones, with a peak GPU computing capability increased by up to 300% compared to the previous generation [5][17]. - The N1 chip supports Wi-Fi 7, Bluetooth 6, and Thread technology, improving the performance and reliability of features like personal hotspots and AirDrop [11][20]. - The C1X modem offers speeds up to 2 times faster than its predecessor, with a 30% reduction in overall power consumption, making it the most power-efficient modem in an iPhone [11][24]. Group 2: Performance Enhancements - The A19 Pro chip, when paired with Apple's custom-designed heat dissipation system, enhances sustained performance by up to 40% compared to the previous generation, suitable for gaming and video editing [17][18]. - The new GPU architecture includes a dedicated neural accelerator in each core, enabling advanced graphics processing and AI model execution [18][26]. - The integration of the N1 and C1X chips with the A19 Pro allows the iPhone Air to achieve professional-grade performance in a lighter and thinner design [26][28]. Group 3: Strategic Direction - Apple aims to reduce reliance on third-party components, such as Qualcomm's modem, by developing its own chips, which is part of a broader strategy to enhance product functionality and efficiency [27][28]. - The company plans to eventually integrate its modem and processor into a single chip, although this may take several years to achieve [27].

Core Viewpoint - ASML, a Dutch lithography giant, has invested €1.3 billion in French AI startup Mistral AI, becoming its largest shareholder, aiming to integrate generative AI technology into its operations to enhance lithography machine development and manufacturing efficiency [1][5][9] Investment and Strategic Partnership - The investment allows ASML to acquire an 11% stake in Mistral AI and aims to develop AI-enabled solutions for its extensive customer base, with ASML's CFO Roger Dassen joining Mistral's board [5][9] - Morgan Stanley views the investment as having a neutral short-term impact on ASML's earnings expectations, while UBS and Wolfe Research maintain a bullish outlook, raising ASML's target price significantly [2][4][10] Market Reactions and Stock Performance - ASML's stock has seen a notable increase, trading around €685 per share in Europe and $805 per share in the US, driven by the High-NA lithography machine deployment and the Mistral AI investment [2][4] - Morgan Stanley maintains a "hold" rating with a target price of €600, while UBS has raised its target price from €660 to €750, reflecting differing market sentiments [2][4][10] Future Growth Prospects - UBS forecasts a compound annual growth rate of 20% for ASML's earnings per share from 2026 to 2030, supported by strong financial fundamentals and demand for High-NA EUV lithography machines from major semiconductor manufacturers [10][11] - The transition of High-NA technology from laboratory validation to commercial deployment is expected to significantly boost ASML's long-term growth prospects [9][12] Technological Advancements - High-NA EUV lithography machines, utilizing a 0.55 NA lens, are crucial for advancing manufacturing processes at 2nm and below, enhancing resolution and efficiency in chip production [12]