该行上调ASMPT纯利预测以反映持续的人工智能需求，并应用27倍2027年预测每股盈利倍数，此乃基 于市场对全球人工智能资本开支，尤其是HBM的高涨情绪。该行维持对ASMPT"买入"评级，目标价由 98港元上调至133港元。 中银国际发表研报指，ASMPT公布的2025年第四季业绩稳健胜预期，受蓬勃的AIDC(人工智能数据中 心)需求及附属公司AAMI出售的一次性收益带动，尽管不利的产品组合部分限制了利润率扩张。管理层 对2026年第一季的强劲指引，包括订单、收入及SEMI毛利率，均指向人工智能-HBM及CoWoS-TCB(利 用热压技术来组装的高阶AI芯片封装技术)需求链将持续至2028年。 ...

Core Insights - The article discusses the increasing demand for AI Data Centers (AIDC) driven by the exponential growth in computing power requirements due to generative AI advancements and supportive government policies like "East Data West Computing" [5][32] - Major tech companies are ramping up investments in AI infrastructure, with ByteDance planning to increase its capital expenditure to approximately 160 billion RMB in 2026, while Alibaba aims to invest over 380 billion RMB in technology R&D and infrastructure over the next three years [7][34] - The article highlights the penetration of AIDC into traditional industries, evidenced by significant procurement projects such as China Mobile's purchase of 7,499 AI servers for 2025-2026 [8][35] AIDC Types and Characteristics - AIDC is categorized into three types: General Data Centers, Intelligent Computing Data Centers (AIDC), and Supercomputing Data Centers, each serving different computational needs [4][30] - General Data Centers focus on traditional data storage and management using CPU servers, while AIDC leverages AI chips like GPUs for large-scale model training, and Supercomputing Data Centers support advanced scientific research [4][30] Five-Layer Cake Theory - NVIDIA's CEO proposed a "Five-Layer Cake" structure for AI infrastructure, which includes Energy Layer, Chip and Computing Layer, Infrastructure Layer, AI Model Layer, and Application Layer [10][37] - The Energy Layer is crucial for providing stable power to AIDC, while the Chip and Computing Layer focuses on high-performance hardware [11][39] - The Infrastructure Layer integrates energy and chip resources to deliver intelligent computing services, and the AI Model Layer is essential for developing models that drive AI applications [13][41] Industry Ecosystem and Opportunities - The AIDC industry's growth is a result of the synergy between computing power demand and technological advancements, benefiting various sectors [18][45] - The transition to high-voltage and direct current power systems is becoming mainstream, with NVIDIA introducing an 800V DC power architecture to meet the power demands of next-gen AI facilities [19][46] - Liquid cooling systems are gaining traction due to the high power consumption of AI servers, leading to increased market demand for cooling technologies [20][47] Domestic AI Chip Market - The domestic AI chip market is diversifying, with multiple brands achieving significant sales volumes, indicating a shift from technology development to large-scale delivery [25][52] - The price range for domestic AI inference chips is between 30,000 to 200,000 RMB, with a notable increase in production expected as manufacturing capacity improves [25][52] Conclusion - The article emphasizes that while China has advantages in energy resources and computing infrastructure, breakthroughs in high-end chip development and core technology innovation are still needed [26][53] - The ultimate winners in the AI industry will be those who can integrate full-stack technologies and foster collaborative industrial advancements [26][53]

Core Viewpoint - Century Internet's stock price reached a new high on February 12, 2026, driven by strong financial performance, proactive business expansion, and asset revitalization efforts [1] Financial Performance - For the third quarter of 2025, the company reported total revenue growth of 21.7% year-on-year, reaching 2.58 billion RMB, with adjusted EBITDA increasing by 27.5% to 758 million RMB [2] - The company raised its full-year guidance, expecting total revenue growth of 16% to 19% and adjusted EBITDA growth of 24% to 26% for 2025 [2] - In the first quarter of 2025, total revenue also achieved an 18.3% year-on-year growth, providing fundamental support for the stock price [2] Business Development - In the third quarter, the company secured three base-type IDC orders with a total capacity of 63 MW, and in the fourth quarter, it received a significant order of 32 MW from an internet client [3] - As a leading player in the AIDC (Artificial Intelligence Data Center) sector, the company is benefiting from market demand driven by AI acceleration [3] - As of September 30, 2025, the company's base-type IDC installed capacity increased by 13.8% quarter-on-quarter, with an installation rate of 74.3%, indicating strong operational efficiency [3] Project Progress - In January 2026, two IDC projects of Century Internet received approval for private REITs (asset-backed special plans), with a total proposed issuance amount of approximately 10.759 billion RMB [4] - This initiative helps the company revitalize heavy assets and recoup funds for new project investments, alleviating financial pressure and viewed positively by the market [4] Industry Policy and Environment - Analysts from institutions like Goldman Sachs indicate that the demand for data center tenders from major internet companies may be larger and come sooner than expected, leading to significant capital expenditure increases by leading operators in 2026 to accelerate capacity delivery [5] - This creates a favorable industry environment for leading IDC companies like Century Internet [5]

Investment Rating - The report does not explicitly state an investment rating for the industry [1]. Core Insights - North American data centers (DC) are experiencing rapid growth, with operational capacity reaching 25.6 GW and projects under construction hitting a record 12.1 GW, of which 88.5% is pre-committed [10][12]. - The trend in AI data center (AIDC) capacity is shifting towards centralized training, which offers advantages in management and synchronization, with expectations for single data center capacities to approach 1 GW [11][12]. - Power availability is critical for site selection, particularly in the PJM region, where new AIDC projects may face power interconnection delays of 4-7 years [12][13]. - AIDC projects are increasingly adopting self-supplied primary generation to address power needs while awaiting grid access, primarily using gas turbines [13][14]. - Backup power solutions are evolving, with a shift from diesel generators to gas turbines due to regulatory constraints on diesel use [14]. - Energy storage systems are being integrated to smooth out power demand fluctuations during training and inference phases, enhancing overall supply reliability [15]. Summary by Sections AIDC Capacity Development Trends - Centralized training is favored over decentralized architectures due to its efficiency in management and synchronization, with a projected increase in single-site compute density [11][12]. Power Availability and Site Selection - The PJM region is the largest ISO for data centers, but lengthy power interconnection timelines are pushing AIDC projects to other ISOs like MISO and ERCOT [12]. Changes in Power Interconnection Approaches - New AIDC projects are increasingly utilizing self-supplied primary generation to mitigate delays in grid access, primarily through gas turbines [13]. Trends in Backup Power - The industry is moving towards greater use of gas turbines for backup power due to regulatory challenges associated with diesel generators [14]. Energy Storage Integration - Energy storage systems are being utilized to manage power demand fluctuations, improving reliability and resilience of AIDC operations [15].

Core Insights - The essence of the competition in computing power has evolved beyond just chips and servers to a deeper contest of technological foundation and operational capabilities [1] - Runze Technology has made significant strides in the AI computing power sector by focusing on technological innovation, green low-carbon initiatives, and talent strategies [1][2] Group 1: Company Overview - Runze Technology was founded in 2009, recognizing the impending demand for data storage due to the internet explosion, and strategically located in Langfang to serve major internet companies while reducing operational costs by over 20% compared to Beijing [2] - The company’s flagship project, Runze International Information Port, covers an area of 2.62 million square meters with a total investment of 9.8 billion yuan, equivalent to two Chicago data centers [2] - Runze Technology transitioned from traditional data centers to intelligent computing infrastructure, achieving a significant milestone with its listing on the ChiNext board in 2022 [2][3] Group 2: Financial Performance - In 2023, Runze Technology reported a revenue of 4.351 billion yuan, a year-on-year increase of 60.27%, with a net profit of 1.762 billion yuan [3] - For 2024, the company expects stable operations with projected revenue of 4.365 billion yuan and a net profit of 1.79 billion yuan [3] - In the first three quarters of 2025, revenue reached 3.977 billion yuan, reflecting a year-on-year growth of 15.05%, while net profit surged to 4.704 billion yuan, a remarkable increase of 210.74% [3] Group 3: Business Segments - Runze Technology operates two main business segments: Internet Data Centers (IDC) and Artificial Intelligence Data Centers (AIDC), with AIDC being the core growth driver [3] - The company has established seven AIDC computing power clusters across key regions, planning for 320,000 cabinets to create an integrated computing center system [3] Group 4: Talent Development - The company has built a highly specialized operational team of over 1,000 members, adhering to international standards such as ITIL and Uptime, to provide reliable operational services [4] - Runze Technology has implemented a performance-based compensation system and established a dual career path for technical personnel, promoting innovation and talent retention [5] Group 5: Technological Innovation - To address the dual challenges of computing power escalation and green transformation, Runze Technology focuses on technological advancements, including the development of liquid cooling technology to enhance cooling efficiency and reduce energy consumption [6] - The company’s intelligent low-carbon cooling system has achieved a Power Usage Effectiveness (PUE) of 1.15, positioning it as an industry leader [6] Group 6: Future Projects - The Runze Artificial Intelligence Application Center project is in its final stages, expected to be operational by the end of 2025, capable of supporting over 100,000 computing clusters for complex AI model training [7] - This project aims to become a national-scale intelligent computing center, providing comprehensive support for various AI technologies [7]

Core Insights - AI is a key thread connecting the "Five Major Articles" of China's strategic development, emphasizing the importance of aligning investment strategies with national policies [1][8] - The capital market is experiencing a significant increase in activity, with funds actively seeking investment opportunities closely related to AI [3][4] - The transition of AI from experimental technology to value creation is recognized as a core driver of the "technology + industry" revolution [6][7] Capital Market Dynamics - Since the implementation of the "9·24" policy in 2024, the Chinese capital market has seen two notable phases of upward movement, driven by policy benefits and subsequent market stabilization [4] - M1 money supply increased by 6% in August 2025 compared to a 7.3% decline in the same month of 2024, indicating a shift in asset allocation towards equities [4] - In July, household deposits decreased by 1.1 trillion yuan, while securities and fund accounts increased by 2.14 trillion yuan, reflecting a trend of moving funds from savings to equity markets [4] AI Industry Insights - The AI industry is witnessing a period of significant breakthroughs, with companies like Cambrian Technologies demonstrating the potential for independent development in China's AI sector [7] - The market is expected to see a revaluation of domestic AI-related assets as the gap between China's AI capabilities and global leaders narrows [7] - AI is being integrated across various sectors, including manufacturing, education, finance, and healthcare, indicating a broad transformation driven by AI technologies [7][8] Investment Strategies - Investment institutions are encouraged to align with national strategies and focus on the intersections of AI and the "Five Major Articles" [8] - Companies are advised to invest in AI-related core technologies and software, particularly in early-stage ventures that can leverage AI for operational efficiency [8][9] - The reduction in costs and increased accessibility of AI tools are opening up investment opportunities across industries, making AI integration a key factor in future investment value [9]

Core Viewpoint - AI is a key thread connecting the "Five Major Articles" in China's capital market, representing significant investment opportunities at the intersection of policy, technology, and demand [1][10]. Group 1: Capital Market Dynamics - The Chinese capital market is experiencing a notable increase in activity, with funds seeking investment directions closely related to AI [3][4]. - Since the implementation of the "9·24" policy in 2024, the capital market has seen two significant phases of upward movement, driven by policy benefits and subsequent AI technology breakthroughs [4]. - As of August 2025, M1 (narrow money supply) grew by 6% year-on-year, contrasting with a 7.3% decline in the same period of 2024, indicating a shift in liquidity towards capital markets [4]. - In July, household deposits decreased by 1.1 trillion yuan, while securities and fund accounts increased by 2.14 trillion yuan, showing a trend of reallocating funds from banks to equity assets [4][5]. - The A-share market has seen daily trading volumes exceed 2 trillion yuan, signaling a peak in market activity for the year [4]. Group 2: AI Industry Insights - The AI industry is transitioning from technical exploration to value creation, becoming a core driver of the "technology + industry" revolution [7]. - The performance of companies like Cambrian Technology indicates that China's AI industry is on a path of independent development, narrowing the gap with global leaders [8]. - AI is being integrated across various sectors, including manufacturing, education, finance, and healthcare, fundamentally transforming traditional industries [8][11]. - The greatest value from AI is expected to emerge from its application layer, where it can help businesses reduce costs and improve efficiency [8]. Group 3: Investment Strategies - Investment institutions should align with national strategies and focus on the intersections of AI and the "Five Major Articles" [9][10]. - 弘毅投资 is actively investing in AI-related sectors, including artificial intelligence data centers, smart energy management, and semiconductor technologies [10]. - The investment approach should prioritize early-stage investments in core technologies related to AI, emphasizing the importance of cash flow generation from AI applications [8][10][11]. - The integration of AI into various industries will determine future investment values, as smaller enterprises gain access to AI technologies previously dominated by larger firms [11].

Core Viewpoint - The article discusses the unprecedented challenges faced by data center infrastructure due to the explosive growth in computing power demand driven by artificial intelligence (AI) technologies, emphasizing the need for a new generation of Artificial Intelligence Data Centers (AIDC) [1][4]. Group 1: AIDC Development and Standards - The first AIDC Industry Development Conference was held in Shanghai, where representatives from various sectors discussed standardization, technology collaboration, and ecosystem synergy, leading to the pre-release of the "AIDC Infrastructure Specification" to guide AIDC construction over the next 2-3 years [2][8]. - The AIDC infrastructure standard aims to address the challenges of planning, delivery, and operation in the rapidly evolving AIDC industry, promoting standardized and scalable development [8][9]. Group 2: Computing Power Demand - The demand for computing power is experiencing explosive growth as AI technology penetrates various industries, becoming a key driver of technological revolution and industrial transformation [4][5]. - The compound annual growth rate (CAGR) for domestic AI servers is over 20%, while the scale of intelligent computing power is growing at over 30%, indicating a need for continuous upgrades in AI infrastructure [5][6]. Group 3: Challenges in AIDC Construction - AIDC construction faces significant challenges related to heat, electricity, and space, as well as a lack of standards and long construction cycles, which complicate the establishment of AI data centers [8][9]. - The industry is currently in a phase of exploration, leading to cost inefficiencies and extended delivery times, with data center construction often taking years, which delays the deployment of AI equipment [8][9]. Group 4: Future Trends and Innovations - By 2035, the total computing power in society is projected to increase by 100,000 times, with disruptive innovations expected in computing architecture, materials, engineering processes, and computing paradigms [5][6]. - The emergence of large models in AI is expected to create numerous new application scenarios, marking a period of rapid application explosion in the AI industry [6][7].

Core Insights - The rapid evolution of artificial intelligence technology has made computing power a core engine driving the digital economy, leading to increased investment in Artificial Intelligence Data Centers (AIDC) by leading companies in the technology and telecommunications sectors [1][2] - The AIDC industry is expected to experience rapid growth over the next two to three years, with significant opportunities emerging as companies collaborate to address challenges and develop industry standards [1][2] Industry Overview - As of June 2025, China is projected to have 10.85 million standard racks in use for computing power centers, with intelligent computing power expected to reach 788 EFLOPS [1] - The share of intelligent computing power in total computing power in China is anticipated to rise from 3% in 2016 to 35% by 2025 [1] - The market size of China's intelligent computing center is expected to reach 135.6 billion yuan by 2025, marking a significant milestone in the AI infrastructure competition [3] Challenges Faced - The AIDC industry faces multiple challenges, including heat, electricity, and space constraints, as well as a lack of standards and long construction cycles [2][5] - The need for standardization and productization in infrastructure is viewed as a crucial path to address these challenges [5] Company Initiatives - Major companies like JD Cloud and iFlytek are accelerating their AIDC construction efforts, with JD Cloud managing 22,000 racks across four self-built data centers [3] - China Mobile is developing an "N+X" intelligent computing center layout, having established two national nodes and twelve regional nodes, integrating its self-developed "Jiutian" model [3][4] Standardization Efforts - The establishment of systematic standards for AIDC infrastructure is critical for efficient and interconnected development, with industry leaders advocating for modularization and standardization of products [5] - The Global Computing Coalition (GCC) is working with industry players to develop standards that will support high-quality AIDC construction and optimization [5]