Group 1 - Nvidia currently holds over 80% of the market value share in the AI server sector, while ASIC AI servers account for approximately 8%-11% [1][3][4] - Major cloud service providers like Meta and Microsoft are planning to deploy their own AI ASIC solutions, with Meta starting in 2026 and Microsoft in 2027, indicating potential growth for cloud ASICs [1][4][10] - The total shipment of AI ASICs is expected to surpass Nvidia's AI GPUs by mid-2026, as more cloud service providers adopt these solutions [4][10] Group 2 - Meta's MTIA AI server project is anticipated to be a significant milestone in 2026, with plans for large-scale deployment [2][13] - Meta aims to produce 1.5 million units of MTIA V1 and V1.5 by the end of 2026, with a production ratio of 1:2 between the two versions [21][22] - The MTIA V1.5 ASIC is expected to have a larger package size and more advanced specifications compared to V1, which may pose challenges during mass production [23][19] Group 3 - Companies like Quanta, Unimicron, and Bizlink are identified as potential beneficiaries of Meta's MTIA project due to their roles in manufacturing and supplying critical components [24][25][26] - Quanta is responsible for the design and assembly of MTIA V1 and V1.5, while Unimicron is expected to supply key substrates for Meta and AWS ASICs [24][25] - Bizlink, as a leading active cable supplier, is poised to benefit from the scaling and upgrading connections in Meta's server designs [26]

Core Insights - The article discusses Meta's ambitious plans to challenge NVIDIA's dominance in the AI server market with its MTIA series of products, aiming for significant production volumes and advanced technology [3][5]. Group 1: Meta's MTIA Series - Meta is developing the MTIA T-V1, T-V1.5, and T-V2 models, with T-V1 and T-V1.5 expected to have a combined shipment target of 1-1.5 million units by the end of 2026, with T-V1.5 making up a larger share [5][4]. - The MTIA T-V1.5 will feature a doubled intermediary layer size and a PCB with up to 40 layers, while T-V2 is expected to utilize larger CoWoS packaging and require liquid cooling systems with power systems exceeding 170KW [4][5]. Group 2: Market Dynamics - In 2026, the shipment volume of AI ASICs is projected to surpass that of GPGPUs, with NVIDIA currently holding over 80% of the AI server market value share, while ASIC AI servers account for approximately 8-11% [6][7]. - By 2025, the combined shipment of Google TPU and AWS Trainium 2 is expected to reach 40-60% of NVIDIA's GPU shipments, indicating a shift in market dynamics [7]. Group 3: Cost and Technical Considerations - The article highlights a divergence in views regarding the necessity and cost of ASICs, with some large cloud service providers (CSPs) advocating for ASICs to bridge technology gaps despite higher BOM costs, while NVIDIA's CEO suggests that most ASIC projects may be canceled due to economic viability concerns [9][10]. - The high costs associated with ASICs stem from the need for larger CoWoS packaging and high-spec PCBs, which are essential for performance compensation and thermal management [11][12]. Group 4: PCB Market Potential - The article notes that the market for ASIC and GPU PCBs is expected to double in value, with a projected growth rate of 65% next year, indicating a significant opportunity for PCB suppliers [16]. - Companies like Huada Empyrean (WUS) are highlighted as potential beneficiaries of this growth, providing advanced PCBs for Meta's MTIA [17][19].

Group 1 - The report maintains an overweight rating on the computer industry, highlighting the rise of AI and domestic chip manufacturers as key investment themes [3][4] - AI applications and computing power are expected to see significant growth, with AI Agents poised for a breakthrough year in 2025, particularly in enterprise services and SaaS sectors [4][5] - The domestic AI server market is projected to grow significantly, with a forecasted market size exceeding $100 billion by 2029, driven by increasing demand for AI computing power [5][4] Group 2 - The report emphasizes the new opportunities in the domestic software industry, particularly in core software areas where the localization rate is currently low, indicating a potential peak replacement period [4][5] - The report notes that the increasing external uncertainties and government support are accelerating the development of domestic software, with significant room for growth in database and operating system sectors [5][4] - The report identifies that the domestic database market is expected to see accelerated penetration in various industries as the 2027 full replacement deadline approaches [5][4] Group 3 - The report discusses the competitive landscape of AI models, noting that the performance gap between domestic and international models is narrowing, with significant advancements in domestic open-source models [5][28] - The report highlights the emergence of AI Agents as a new paradigm in human-machine collaboration, with the potential to transform efficiency in various sectors [31][32] - The report outlines the parallel development of MCP and A2A protocols, which are expected to enhance the AI Agent ecosystem by facilitating interoperability and data sharing [38][39]

Core Insights - Apple is aggressively pursuing the AI glasses market, with a modular design that allows users to customize components such as batteries and frames, enhancing personalization and usability [1][2] - The anticipated launch of Apple's AI glasses is expected in 2026, with Foxconn and Quanta likely to secure manufacturing contracts, while TSMC is expected to produce key chips [1] - The introduction of AI glasses is expected to spark a new wave of market interest, benefiting various Taiwanese stocks associated with AI glasses [1] Modular Design Features - The patented modular design includes detachable support arms that can alleviate pressure on the nose and cheeks, enhancing comfort for users [2] - Users can replace components to improve fit, extend battery life, or personalize style, making the glasses adaptable for various head shapes and activities [2] - The support arms are embedded with essential electronic components, including processors, speakers, and batteries, allowing for a swappable battery feature [2] Technological Innovations - The AI glasses are expected to include a camera, microphone, and speaker, with built-in Siri support for functionalities like phone calls, music playback, instant translation, and navigation [3] - The design incorporates a "liquid glass" concept that creates a semi-transparent effect, allowing digital information to blend seamlessly into the user's view [3][4] - This innovative design aims to provide a more intuitive and natural interaction experience by merging digital content with the physical world [4]

Core Insights - OpenAI has launched its most powerful AI reasoning model, o3-pro, and significantly reduced the price of its flagship model o3 by 80%, indicating a shift towards more accessible AI technology [1][2] - The new pricing strategy is expected to drive demand for AI applications, benefiting companies like Hon Hai, Quanta, Wistron, and Inventec [1] - The o3-pro model is designed for complex logical problems and has outperformed competitors like Google’s Gemini 2.5 Pro and Anthropic’s Claude 4 Opus in key benchmarks [1] Pricing Strategy - OpenAI has drastically reduced the API usage costs for the o3 model from $10 per million input tokens and $40 for output tokens to $2 and $8 respectively [2] - The new o3-pro model is priced at $20 for input and $80 for output per million tokens, marking an 87% reduction compared to the previous o1-pro model [2] Future Outlook - OpenAI's CEO, Altman, predicts that by 2025, AI agents with cognitive abilities will emerge, transforming programming methods, with further advancements expected in 2026 and 2027 [2] - Altman emphasizes that the era of "intelligent technology being immeasurably cheap" is approaching, suggesting rapid advancements in AI capabilities [3]

全球关税战方兴未艾，对生成式AI投资回报过慢的质疑，以及美国持续扩大对中国大陆AI运算力的技 术、产品封锁，皆未影响主要业者继续大举投入建置AI资料中心，大型语言模型(LLM)效能与应用持续 发展，仍吸引业者不断投资并采购AI服务器。DIGITIMES预估，2025年全球AI服务器出货将达181万 台，而搭载高频宽存储器(High Bandwidth Memory；HBM)的高阶AI服务器出货则将突破百万台大关， 较2024年成长4成。 由于GB200/GB300机种出货占比将从2024年接近0%，大幅成长至近4成，与GB200主板、整机、机柜相 关的台系代工厂皆将显著受惠，其中，鸿海（2317）2025年在主机板(L6)产量的排名将由2024年的第四 上升至第一。在整机系统生产(L10)方面，长期与Google合作的Celestica占比仍将维持第一，但广达 （2382）、鸿海、纬创（3231）的占比亦将大幅提升。 DIGITIMES指出，2025年高阶AI服务器采购量成长最快的预估为甲骨文(Oracle)，主因甲骨文为OpenAI 的星际之门计划(Stargate Project)合作业者中最积极者之一； ...

Group 1 - The "Made in America" strategy promoted by President Trump is set to achieve its first victory with Foxconn expanding AI server production in the U.S. [1] - Foxconn plans to shift a significant portion of its GPU modules and motherboards production to the U.S., currently over 90% of these components are manufactured in Taiwan [1][2] - The expansion in Texas is crucial for Foxconn to enhance supply chain resilience amid geopolitical uncertainties [1] Group 2 - Foxconn and Wistron are the only two suppliers producing critical components for NVIDIA's AI servers, which require substantial engineering resources [2] - Over 1,000 engineers at Foxconn are dedicated to supporting NVIDIA's AI system development, covering various production stages [3] - Other suppliers like TSMC and SK Hynix are also expanding their operations in the U.S. to support the AI server supply chain [3][4]

Core Viewpoint - The report from Morgan Stanley highlights a significant increase in the global production of GB200 NVL72 racks, driven by the surging demand for AI computing, particularly in cloud computing and data center sectors [1][2]. Group 1: Production Forecast - The global total production of GB200 NVL72 racks is estimated to reach 2,000 to 2,500 units by May 2025, a notable increase from the previous estimate of 1,000 to 1,500 units in April [1]. - The overall production for the second quarter is expected to reach 5,000 to 6,000 units, indicating a robust supply chain response to market demand [1]. Group 2: Company Performance - Quanta shipped approximately 400 GB200 racks in May, a slight increase from 300 to 400 units in April, with monthly revenue reaching about 160 billion New Taiwan Dollars, a year-on-year increase of 58% [2]. - Wistron demonstrated a strong growth trajectory, shipping around 900 to 1,000 GB200 computing trays in May, a nearly sixfold increase from 150 units in April, with revenue growth of 162%, reaching 208.406 billion New Taiwan Dollars [2]. - Hon Hai shipped nearly 1,000 GB200 racks in May, with a forecast of delivering 3,000 to 4,000 racks in the second quarter, despite some decline in its cloud and networking business due to traditional server shipment slowdowns [2]. Group 3: Market Dynamics - The actual delivery volume of GB200 racks may be lower than the reported shipment figures due to the need for further assembly of Wistron's L10 computing trays into complete L11 racks, which involves additional testing and integration time [3]. - Morgan Stanley ranks the preference for downstream AI server manufacturers as Giga-Byte, Hon Hai, Quanta, Wistron, and Wiwynn, with Giga-Byte being favored for its potential in GPU demand and the server market [3]. - A report from Tianfeng Securities indicates that major hyperscale cloud providers are deploying nearly 1,000 NVL72 cabinets weekly, with the shipment pace continuing to accelerate [3].

Core Viewpoint - The article highlights the emerging potential of Chongqing as a significant player in China's foreign trade landscape, particularly in the context of the ongoing trade dynamics and the city's strategic advantages in logistics, talent, and manufacturing capabilities [1][50]. Group 1: Trade Performance and Growth - In 2024, Chongqing's export value is projected to reach approximately 507.3 billion yuan, ranking first among western cities with a year-on-year growth of 6.2% [1][51]. - The export categories from Chongqing are primarily high-value products such as electromechanical products, automobiles, and motorcycles, which provide greater resilience against tariff fluctuations compared to lower-value goods [8][12]. - The export value of electromechanical products is expected to be around 454.97 billion yuan, accounting for 89.7% of Chongqing's total exports, while automobile exports are projected at 43.13 billion yuan, growing by 30% [12][13]. Group 2: Logistics and Infrastructure - Chongqing benefits from a robust logistics network, including the China-Europe Railway Express and the Western Land-Sea New Corridor, which significantly enhances its export capabilities [18][36]. - In 2024, the value of goods transported via the Western Land-Sea New Corridor is expected to reach 46.7 billion yuan, marking a 67% increase [36]. - The China-Europe Railway Express has expanded its network to connect 125 cities in China and over 200 cities in 25 European countries, facilitating smoother trade routes for Chongqing [34][38]. Group 3: Market Dynamics and Challenges - The article discusses the challenges faced by Chongqing's foreign trade, particularly the need for flexibility in the supply chain to adapt to the growing role of cross-border e-commerce, which saw a transaction value of 68 billion yuan in 2024, up 67% from previous years [22][23]. - The current trade dynamics indicate that high-value products are less affected by tariff changes, allowing companies in Chongqing to maintain stable operations despite fluctuations in the global market [10][11]. - The article emphasizes the importance of integrating small and medium-sized enterprises to enhance production willingness and adapt to the evolving market landscape [23][24]. Group 4: Economic Context and Future Outlook - The article positions Chongqing's foreign trade growth as part of a broader shift in China's economic landscape, moving from an export economy dominated by coastal cities to a more balanced approach that includes western regions [51]. - In 2024, China's total goods trade export value is expected to reach 25.45 trillion yuan, with Chongqing's exports contributing significantly to this growth [47]. - The potential for Chongqing to emerge as a "foreign trade city" is underscored by its strategic advantages in logistics, talent, and manufacturing, which are crucial for its future development in the global trade arena [50][52].

Group 1 - Nvidia's AI server GB200 assembly cooling issues have been resolved, with expectations for significant performance boosts in AI server supply chain companies starting from Q2 [1] - The GB200 server is priced at ten times that of regular servers, indicating a strong potential for revenue growth in the second half of the year [1] - Major suppliers in the GB200 supply chain include TSMC, Hon Hai, Quanta, and others, highlighting a robust ecosystem supporting Nvidia's AI initiatives [1] Group 2 - Hon Hai reported record revenue of NT$641.36 billion in April, reflecting a 16.1% month-over-month increase and a 25.5% year-over-year increase, attributed to the large shipment of GB200 servers [2] - Quanta's executive noted that GB200 shipments began in March, with expectations for continued growth in performance and operations in the second half of the year [2] - Morgan Stanley estimates that GB200 shipments reached 1,500 units in April, a significant increase from 1,000 units in Q1, with Hon Hai being the largest contributor [3]