Production & Capacity - AI6 is expected to start mass production in 2027, with a potential production timeline close to AI5's, aligning with Musk's statements [1] - Tesla aims to participate in the wafer fabrication business at a very low cost, potentially by transferring AI6 orders to Samsung [1] - Musk expresses concerns about future chip supply, even with optimistic projections from suppliers [2] - TSMC assures that chip supply is not an issue if Tesla is willing to pay, although TSMC's capacity can be tight during peak market conditions [2] Strategic Motivations for Tesla's Fab - Geopolitical concerns exist regarding the concentration of semiconductor production in Taiwan [3] - Advanced packaging capacity in the US is expected to be limited, potentially accounting for only 10% of the global capacity by 2030 [3] - Tesla may receive less R&D support and production flexibility from TSMC compared to first-tier clients like Apple and Nvidia [3] - Tesla aims to customize key designs and manufacturing processes to maximize vertical integration benefits [3]

Semiconductor Strategy & Validation - Ming-Chi Kuo's analysis and predictions regarding Elon Musk's semiconductor strategy are validated by Musk's remarks at the latest Tesla shareholder meeting [1] - Musk anticipates transitioning to AI6 within a year of AI5 production, aiming to double performance metrics, aligning with industry projections of AI6 mass production in 2027 [1] Tesla's Chip Production Ambitions - Musk's intention to build Tesla's own chip production plants validates the view that shifting AI6 orders to Samsung was to gain foundry experience at a low cost [2] - Musk is concerned about future chip supply, stating current supplier capacity projections are insufficient [2] - The industry believes TSMC is unlikely to be the primary bottleneck, as TSMC CEO indicated willingness to supply chips if Tesla is willing to pay [2] Geopolitical & Strategic Considerations - Geopolitical concerns, particularly the concentration of advanced node capacity in Taiwan, drive Musk's desire for a Tesla chip production plant [3] - TSMC's advanced-node and advanced-packaging capacity in the U S is expected to remain limited, likely no more than approximately 10% of its global capacity by 2030 [3] - Tesla, as a second-tier customer at TSMC, experiences less priority on R&D support and production flexibility, motivating the move of AI6 to Samsung and the pursuit of its own chip production [3] Integration & Customization Advantages - Customizing key design and manufacturing segments, particularly chip production, enables a highly integrated final product and maximizes the benefits of vertical integration for Tesla's cutting-edge technologies [4]

Product & Strategy - NVIDIA is developing a next-generation AI server power strategy, codenamed Kyber, spanning from GPU/rack-level to data-center scale [1]

Industry Focus - The report discusses NVIDIA's AI server power specifications roadmap, moving from GPU/cabinet level to a data center-centric "Kyber" strategy [1] - The article likely focuses on the technical specifications and strategic implications of NVIDIA's next-generation AI server architecture [1] Technological Advancement - The content highlights NVIDIA's shift towards a more holistic, data center-level approach to AI server design and power management [1] - The "Kyber" strategy likely represents a significant advancement in NVIDIA's AI infrastructure offerings [1] Strategic Implications - The report suggests a focus on the evolving power requirements and design considerations for AI infrastructure [1] - The shift towards data center-level solutions indicates a growing demand for scalable and efficient AI computing resources [1]

iPhone Air demand has fallen short of expectations, leading the supply chain to begin scaling back both shipments and production capacity. Most suppliers are expected to reduce capacity by more than 80% by 1Q26, while some components with longer lead times are expected to be discontinued by the end of 2025. This indicates that the existing Pro series and standard models already cover the majority of high-end user demand well, leaving little room to carve out new market segments and positioning — as previous ...

iPhone Air需求低於預期，供應鏈已經開始降低出貨與產能。供應鏈的產能普遍到1Q26會縮減80%以上，部分出貨前置時間較長的零組件，預計在2025年底前停產。這意味著既有的Pro系列與標準版機型，已經很好得涵蓋了大部分的高階用戶需求，很難再找到新的市場區隔與定位 (從mini、Plus到至今Air的嘗試都沒有成功)。 ...

The press release didn’t make it clear that after the first Nvidia Blackwell wafer was produced in the U.S., it would still need to be shipped to Taiwan for CoWoS advanced packaging — only then would the production of the Blackwell chip be considered complete. Two years from now, if advanced packaging could start in the U.S., that would already count as ahead of schedule.https://t.co/ilcl7qbJx9 ...

Supply Chain - Nvidia Blackwell 晶圆在美国生产后仍需运往台湾进行 CoWoS 封装才能完成 [1] - 美国若能在两年后开始先进封装，则进度已超预期 [1]

As I predicted a year ago, M5 chip is the most important hardware upgrade in the new Vision Pro.https://t.co/ywsA8MCWJp https://t.co/ad56Yg111S郭明錤 (Ming-Chi Kuo) (@mingchikuo):AI spatial computing to be the main selling point for new M5-powered Vision Pro; new Apple suppliers like Zhaowei in focushttps://t.co/FPX2dYgmX1 ...

如我一年前所預測，M5晶片是新款Vision Pro最重要的硬體更新https://t.co/ywsA8MCWJp https://t.co/9iSCacafng郭明錤 (Ming-Chi Kuo) (@mingchikuo):AI空間運算為新款M5版本的Vision Pro最大賣點，Apple供應鏈新進者如兆威機電為關注焦點https://t.co/g57WzzEJjq ...