Core Viewpoint - TSMC, as a pure foundry, has established itself as a reliable manufacturing partner for fabless chip design companies, overcoming initial technological gaps and evolving into a dominant player in the semiconductor industry [1][3][8]. Group 1: Early Challenges and Technological Development - TSMC faced significant challenges in its early years, with a technology gap of two process nodes compared to leading manufacturers, which necessitated starting with less advanced orders [1][3]. - By 1991-1992, TSMC had narrowed the technology gap to one process node, leading to an influx of orders from fabless companies [3][8]. - The first significant order came from Intel for the MCU 80C51 chip, which marked a pivotal moment in TSMC's early success [3][4]. Group 2: Strategic Partnerships and Market Position - TSMC's refusal to create a production line identical to Intel's, opting instead to develop its own systems, established its reputation as a qualified foundry [4][5]. - Following Intel's certification, TSMC gained orders from other vertical manufacturers, solidifying its market position [5][7]. - By 2004, TSMC held a 47% market share in the foundry sector, which increased to 61% by 2023, showcasing its growth and dominance [8]. Group 3: Evolution of Clientele and Industry Impact - TSMC's client base evolved from primarily vertical manufacturers to a majority of fabless companies by 2004, reflecting a significant shift in the semiconductor industry [8][9]. - Notable clients included Altera, ATI, and Qualcomm, indicating TSMC's integral role in the semiconductor ecosystem [8][10]. - TSMC's platform model has been crucial for companies like NVIDIA and Qualcomm, enabling them to innovate without competing directly with TSMC [9][17]. Group 4: Technological Advancements and Market Dynamics - TSMC's advancements in process technology below 10nm have driven innovation in high-performance computing and AI applications [19]. - Intel's decision to establish a foundry division and later outsource production to TSMC highlights the industry's shift towards separation of design and manufacturing [19][20]. - TSMC's role as a trusted foundry has allowed it to serve multiple clients, including Apple and AMD, without conflict, reinforcing its position in the market [18][19].

Summary of Key Points from the Conference Call Industry Overview - The focus is on the **Global Semiconductors & Hardware** industry, specifically the **AI server and edge AI** sectors [2][3]. Core Insights and Arguments - **Investment Surge**: Total investment in upcoming and under-construction data centers has reached approximately **US$750 billion**, with major contributions from hyperscalers, neoclouds, and sovereign funds. Expected capex growth for major cloud service providers (CSPs) is **46% YoY in 2025** and **6% in 2026**, reaching **US$386 billion** [3][30]. - **AI Funding**: Primary market AI funding hit a record **US$60 billion** in 1Q25, largely driven by OpenAI's **US$40 billion** financing round, representing **53% of all global startup funding** [25][30]. - **Server Shipments**: High-end GPU server shipments are projected to grow over **50% in 2025**, while ASIC server shipments are expected to comprise nearly **45% of total AI chip shipments** this year [4][37]. - **Financial Performance**: The AI supply chain remains resilient, with companies like Accton, Gold Circuit, and Wiwynn seeing upward revisions in earnings due to stronger-than-expected demand for ASIC servers [6][38]. Important Developments - **Edge AI Innovations**: While edge AI technologies are gaining traction, they are still in the early stages. AI glasses shipments surged by triple digits YoY in 1Q25, indicating competitive dynamics in the market [7]. - **Key Projects to Monitor**: Upcoming developments include monitoring capex guidance from CSPs, progress on mega projects like Stargate, and advancements in TSMC's AP7/AP8 technologies [8][26]. Potential Investment Opportunities - **Chroma and Delta**: Rated as outperform with price targets of **NT$480** and **NT$490** respectively, benefiting from the AI wave [11][12]. - **NVIDIA**: Rated outperform with a price target of **US$185**, capitalizing on the datacenter opportunity [17]. - **Broadcom and Marvell**: Broadcom expects a **US$60-90 billion** serviceable available market (SAM) for AI revenue by FY2027, while Marvell anticipates a **53% CAGR** in its data center total addressable market (TAM) from 2023 to 2028 [53][56]. Other Notable Points - **Market Dynamics**: The AI server market is expected to drive the global server market to nearly **US$400 billion by 2026**, with a significant increase in the mix of AI servers [35][39]. - **Supply Chain Resilience**: Despite market volatility, the AI supply chain has shown resilience, with stock performance of key suppliers indicating a recovery trend [6]. This summary encapsulates the critical insights and developments discussed in the conference call, highlighting the robust growth and investment opportunities within the AI server and semiconductor sectors.

Yipin Cai, CFA +852 2123 2669 yipin.cai@bernsteinsg.com In this preview, we revise our model for TSMC, UMC, Vanguard and Novatek to reflect the FX pressure. Otherwise, the projection for TSMC is now higher on better near-term momentum, improved AI outlook and possible price hike. Rapid FX moves to bring down TSMC"s Taiwan-dollar revenue & margins. Based on the latest FX moves, we assume 32.5 NT$ = 1 USD in 2Q25 and 29 for 3Q25 & onward. This means TSMC's NT$ revenue will have ~5% headwind vs. its guidance a ...

Summary of Key Points from the Conference Call Industry Overview - The AI server market is expected to maintain strong growth in Q3 2025, with companies like Quanta, Wistron, and Hon Hai projecting triple-digit revenue growth for the year. [1][4] - The overall performance of the AI computing industry in Taiwan's stock market for the first half of 2025 was robust, with significant revenue increases for AI server manufacturers. [2] Company-Specific Insights - Quanta's AI server revenue share is projected to rise from 60% in Q1 to 70% in Q2, while Wistron's share is expected to increase to 50%-60%, and Inventec's to 30%-40%. [1][2] - Hon Hai's cloud network business showed strong growth compared to the same period last year. [1][2] Market Trends and Projections - The lithium battery business is expected to see a slowdown in demand in the second half of the year due to earlier-than-expected orders and currency impacts, with overall performance anticipated to remain flat or show slight growth. [5] - NVIDIA's GPU series is fully transitioning to liquid cooling, with significant market share in liquid cooling components for large cloud providers. [6] Component and Supply Chain Developments - The PCB industry is experiencing growth driven by smooth GPU supply, particularly from NVIDIA's new GPU shipments and traditional PC business. [7] - BMC shipments in the first half of the year were primarily driven by AI servers, with general servers also benefiting from AI applications. [8] Advanced Technology and Future Plans - TSMC is expected to see a transformation in its advanced packaging technology, with new facilities in the U.S. projected to begin construction in 2028. [3][9] - TSMC's AI chip business revenue is forecasted to double by 2025, with a compound annual growth rate of approximately 45% over the next five years. [3][9] Investment Recommendations - Recommended companies across various segments include: - AI Processors: Haiguang Information, Cambricon - Computing Rental: Xiechuang Data, Hongjing Technology, Youfang Technology, Yingfeng Environment - Server Systems: Industrial Fulian, Zhongke Shuguang, Inspur Information, Huqin Technology - Core Components: - PCB: Shenghong Technology, Hude Electronics - Optical Modules: Jinyisheng, Tianfu Communication, Zhongji Xuchuang - Cooling: Shuguang Shuchuang, Feirongda - Copper Connections: Huafeng Technology - CPU: Haiguang Information, Longxin Zhongke [10][11]

Core Viewpoint - Taiwan Semiconductor Manufacturing Company (TSMC) is a crucial player in the AI chip manufacturing space, serving as a reliable partner for AI chipmakers despite not designing chips itself [1][7]. Group 1: TSMC's Market Position - TSMC is recognized as the world's most advanced semiconductor foundry, with major clients including Nvidia, AMD, Broadcom, and Apple, giving it unmatched scale and technological leadership [2]. - The company has a significant market share lead in the advanced node market, outperforming competitors like Intel and Samsung [2]. Group 2: Revenue and Growth - In Q1, TSMC's revenue increased by 35% to $25.5 billion, driven by high-performance computing (HPC) growth, and preliminary Q2 revenue is estimated to have risen by 39% to $31.9 billion [4]. - Chips manufactured on 7nm and smaller nodes accounted for 73% of TSMC's revenue in Q1, up from 65% the previous year, with 3nm nodes contributing 22% [3]. Group 3: Pricing Power and Margins - TSMC's strong pricing power is evident as it raises prices to counteract margin dilution from new fabs, with gross margin rising by 190 basis points to 58.8% in Q1 [4][5]. - The company plans to increase AI chip prices, with Arizona-made chips potentially commanding a 30% premium [5]. Group 4: Future Growth Prospects - TSMC anticipates AI-related revenue to grow at a mid-40% compounded average growth rate (CAGR) over the next five years starting in 2024 [8]. - The company is also positioned to benefit from the growth of autonomous driving and robotaxis, which will require advanced chips [9]. Group 5: Investment Appeal - TSMC is viewed as a safe investment in the AI semiconductor space, as it provides manufacturing services to all major players without needing to bet on a single chipmaker [11]. - The stock is attractively valued, trading at a forward P/E ratio of 24 based on 2025 estimates and a PEG ratio of less than 0.7, indicating it is undervalued [12].

Investment Rating - The report does not explicitly state an investment rating for the industry Core Insights - The AI-driven demand for high-performance chips is a key growth driver for TSMC, with FY25Q2 revenue reaching NT$933.8 billion (approximately USD 31.9 billion), exceeding market expectations and previous guidance [14][19] - The global PC device sales are projected to increase by 6.5% year-on-year in Q2 2025, reaching approximately 68 million units, with Apple showing the strongest performance [21] - MediaTek reported a June revenue of NT$56.4 billion, a year-on-year increase of 31%, indicating strong growth potential in the AI PC market [27][30] Summary by Sections AI Developments - Moonshot AI launched and open-sourced the Kimi K2 model, which features 32 billion active parameters and aims to enhance agent applications in China [12][7] - Google's Veo AI has integrated a new image-to-video feature into its Gemini application, available in over 150 countries [13] TSMC Performance - TSMC's strong Q2 performance reflects ongoing high demand for advanced processes and packaging capacity driven by AI [14] - Key topics for TSMC's upcoming earnings call include AI demand sustainability, non-AI customer demand changes, and the impact of currency fluctuations on profit margins [19][20] PC Market Dynamics - According to IDC, global PC sales are expected to continue their upward trend, with Apple leading in growth [21][23] - The report highlights the competitive landscape in the AI chip market, particularly between MediaTek and Qualcomm [27][30]

Core Viewpoint - The article discusses the recent developments in trade negotiations, the impact of tariffs on inflation, and the upcoming economic data releases that could influence monetary policy decisions in the US and Europe [1][3][6]. Trade Negotiations and Tariffs - President Trump announced a new round of reciprocal tariffs on several trade partners, leading to a decline in US stock markets, with the Dow Jones down 1.02% and the S&P 500 down 0.31% for the week [1]. - The deadline for the new tariffs to take effect is August 1, and investors are awaiting further news on trade negotiations [1][3]. - The EU is attempting to reach a trade agreement with the US, with concerns that US tariffs could disrupt transatlantic supply chains [6]. Economic Data and Monetary Policy - The upcoming US inflation data, particularly the Consumer Price Index (CPI) for June, is expected to influence the Federal Reserve's decision on potential interest rate cuts later this year [3]. - The Federal Reserve's June meeting minutes indicated concerns about inflation pressures from tariffs, but a belief that a rate cut may occur later in the year [3]. - In the UK, inflation has risen, with the CPI at 3.4% in May, and expectations for a potential rate cut by the Bank of England if economic data continues to underperform [7]. Commodity Markets - Oil prices have stabilized, with WTI crude oil rising 2.16% to $68.45 per barrel and Brent crude oil increasing 3.02% to $70.36 per barrel, amid concerns over summer supply and demand [4]. - Gold prices have also rebounded, with COMEX gold rising 0.73% to $3356 per ounce, as investors seek safe-haven assets amid trade uncertainties [5]. Upcoming Economic Indicators - Key economic indicators to watch include the US PPI, industrial production, and retail sales data, as well as consumer sentiment surveys [3][9]. - In Europe, the ZEW economic sentiment index for Germany will be released, reflecting the economic outlook amid trade tensions [6].

Industry Overview - The AI market is projected to grow at a rate of 26% per year, reaching $1 trillion by 2031, driven by investments in autonomous technology, machine learning, and natural language processing [1] - The chip industry has experienced decades of growth, with Taiwan Semiconductor Manufacturing Company (TSMC) being a key player benefiting from strong demand for advanced chips [4] Company: Taiwan Semiconductor Manufacturing (TSMC) - TSMC is the leader in chip manufacturing for various applications, including smartphones and data centers [3] - The company reported a revenue growth of 35% year over year in the first quarter, with earnings jumping 53% compared to the previous year [4] - Although AI chips currently represent a small portion of TSMC's business, revenue from AI is expected to grow at a mid-40s percentage on a compound annual basis through 2029 [5] - TSMC manufactured over 11,000 chip products for more than 500 customers last year and has the capacity to produce about 17 million 12-inch silicon wafers annually [6] - The stock trades at a forward price-to-earnings ratio (P/E) of 24, with expectations of revenue and earnings growth at an annualized rate of 17% through 2029 [6] - TSMC's stock is projected to potentially double in five years, aligning with the company's earnings growth [7] Company: Oracle - Oracle is a leading cloud database company that has seen its stock rise significantly due to strong financial results and growing demand in cloud computing and AI workloads [9] - The company's total revenue grew by 8% year over year last quarter, but its cloud infrastructure business reported a 52% increase in revenue [10] - Oracle's cloud infrastructure segment, which makes up less than 20% of total revenue, is experiencing demand that exceeds supply, with a notable order for all available cloud capacity [10] - The company is part of the Stargate Project, which aims to invest $500 billion by 2030 to develop AI infrastructure in the U.S., alongside major tech firms [11] - Management anticipates a 70% increase in cloud infrastructure revenue in fiscal 2026, up from 50% in fiscal 2025, with total revenue projected to rise 15% on a compound annual basis through fiscal 2030 [12] - Oracle's forward P/E ratio is 32, which is considered fair given the growth opportunities ahead [13]

Investment Rating - The report maintains a positive outlook on the AI computing sector, indicating a sustained high level of prosperity in the industry [2]. Core Insights - The AI sector is experiencing robust growth, with multiple suppliers such as Xinxing, Taiko, and Wistron achieving record monthly revenues. For instance, Xinxing's revenue in June 2025 increased by 55% year-on-year, and further growth is expected in Q3 due to strong demand from Nvidia's GB series [2][6]. - The mature process segment is stabilizing, with companies like UMC and World Advanced reporting revenue growth of 3% and 6% respectively in Q2 2025. UMC anticipates a 5-7% increase in wafer shipments due to rising demand [2][14]. - The memory sector is recovering from production cuts and inventory destocking, with expectations of price increases in the second half of 2025. For example, Nanya Technology's revenue in June reached a near 35-month high [2][15]. - The logic chip segment shows stable demand, with MediaTek reporting an 18% year-on-year revenue increase in Q2 2025, and a strong market position in mobile application processors [2][16]. - Passive components are also benefiting from AI demand, with Yageo achieving record revenue growth of 11% year-on-year in June 2025 [2][18]. Summary by Sections AI Sector - The AI sector is characterized by high growth, with significant revenue increases across various suppliers, particularly in the server management chip and testing equipment segments [2][6][11]. Mature Process - Companies in the mature process segment are seeking differentiated growth strategies, with stable revenue growth reported by UMC and World Advanced [14]. Memory Sector - The memory sector is expected to see price recovery due to seasonal demand and inventory adjustments, with notable revenue increases from major players [15]. Logic Chips - The logic chip market remains stable, with MediaTek leading in market share and expanding into new areas such as AI ASICs and automotive chips [16]. Passive Components - The demand for passive components is driven by AI applications, leading to significant revenue growth for companies like Yageo [18].

Core Viewpoint - The article discusses the significant shift in the automotive MCU market with the introduction of new embedded storage technologies like PCM and MRAM, moving away from traditional embedded Flash technology. This transition is seen as a strategic move that will have a profound impact on the MCU ecosystem [1][3]. New Storage Pathways - Major MCU manufacturers such as ST, NXP, and Renesas are launching new automotive MCU products featuring advanced embedded storage technologies, indicating a shift from traditional 40nm processes to more advanced nodes like 22nm and 16nm [2]. - The evolution of MCUs is characterized by increased integration of AI acceleration, security units, and wireless modules, positioning them as central components in automotive applications [2]. Embedded Storage Technology Revolution - The rise of embedded non-volatile memory (eNVM) technologies is crucial for addressing the challenges posed by the complexity of software-defined vehicles (SDVs) and the increasing demands for storage space and read/write performance [3]. - Traditional Flash memory is becoming inadequate in terms of density, speed, power consumption, and durability, making new storage solutions essential for MCU advancement [3]. ST's Adoption of PCM - ST has introduced the Stellar series of automotive MCUs featuring phase change memory (PCM), which offers significant advantages over traditional storage technologies [5][6]. - The Stellar xMemory technology is designed to simplify the development process for automotive manufacturers by reducing the need for multiple memory options and associated costs [7][9]. NXP and Renesas Embrace MRAM - NXP has launched the S32K5 series, the first automotive MCU based on 16nm FinFET technology with integrated MRAM, enhancing the performance and flexibility of ECU programming [10]. - Renesas has also released a new MCU with MRAM, emphasizing high durability, data retention, and low power consumption, further showcasing the advantages of MRAM technology [11]. TSMC's Dual Focus on MRAM and RRAM - TSMC is advancing both MRAM and RRAM technologies, aiming to replace traditional eFlash in more advanced process nodes due to the limitations faced by eFlash technology [15]. - TSMC has achieved mass production of RRAM at various nodes and is actively developing MRAM for automotive applications, indicating a strong commitment to new storage technologies [15][16]. Integration of Storage and Computing - The article highlights a trend towards "storage-computing integration," where new storage technologies like PCM and MRAM are not just replacements but catalysts for MCU architecture transformation [19]. - The merging of storage and computing functions is becoming increasingly important in the context of AI, edge computing, and the growing complexity of computational tasks [21]. Conclusion - The MCU landscape is evolving from a focus on basic control systems to a more integrated approach where storage plays a critical role in computing architecture, driven by advancements in embedded storage technologies [23]. - This transformation presents both challenges and opportunities for domestic MCU manufacturers, who must adapt to the rapidly changing technological landscape [23].