Core Viewpoint - TSMC is significantly expanding its CoWoS capacity, with projections indicating a rise from 70k wpm at the end of 2025 to 100-105k wpm by the end of 2026, and further exceeding 130k wpm by 2027, showcasing a growth rate that outpaces the industry average [1][2]. Capacity Expansion - TSMC's CoWoS capacity will reach 675k wafers in 2025, 1.08 million wafers in 2026 (a 60% year-on-year increase), and 1.43 million wafers in 2027 (a 31% year-on-year increase) [1]. - The expansion is concentrated in specific factories, with the Tainan AP8 factory expected to contribute approximately 30k wpm by the end of 2026, primarily serving high-end chips for NVIDIA and AMD [2]. Utilization Rates - Due to order matching issues with NVIDIA, CoWoS utilization is expected to drop to around 90% from Q4 2025 to Q1 2026, with some capacity expansion plans delayed from Q2 to Q3 2026. However, utilization is projected to return to full capacity in the second half of 2026 with the mass production of new projects [4]. Customer Allocation - In 2026, NVIDIA is projected to occupy 50.1% of CoWoS capacity, down from 51.4% in 2025, with an allocation of approximately 541k wafers [5][6]. - AMD's CoWoS capacity is expected to grow from 52k wafers in 2025 to 99k wafers in 2026, while Broadcom's capacity is projected to reach 187k wafers, benefiting from the production of Google TPU and Meta V3 ASIC [5][6]. Technology Developments - TSMC is focusing on advanced packaging technologies such as CoPoS and WMCM, with CoPoS expected to be commercially available by the end of 2028, while WMCM is set for mass production in Q2 2026 [11][14]. - CoPoS technology offers higher yield efficiency and lower costs compared to CoWoS, while WMCM is positioned as a cost-effective solution for mid-range markets [12][14]. Supply Chain and Global Strategy - TSMC plans to outsource CoWoS backend processes to ASE/SPIL, which is expected to generate significant revenue growth for these companies [15]. - TSMC's aggressive investment strategy in the U.S. aims to establish advanced packaging facilities, enhancing local supply chain capabilities and addressing global supply chain restructuring [15]. AI Business Contribution - AI-related revenue for TSMC is projected to increase from 6% in 2023 to 35% in 2026, with front-end wafer revenue at $45.162 billion and CoWoS backend revenue at $6.273 billion, becoming a core growth driver [16].

Core Viewpoint - Microsoft is adjusting its internal AI chip development strategy to focus on less aggressive designs by 2028, aiming to overcome delays in development while maintaining competitiveness against Nvidia [2][4]. Group 1: Development Delays and Strategic Adjustments - Microsoft has faced challenges in developing its second and third-generation AI chips, leading to a strategic shift towards more pragmatic and iterative designs [2][4]. - The Maia 200 chip's release has been postponed from 2025 to 2026, while the new Maia 280 chip is expected to provide a 20% to 30% performance advantage per watt over Nvidia's 2027 chip [2][4][5]. - The company acknowledges that designing a new high-performance chip from scratch each year is not feasible, prompting a reduction in design complexity and an extension of development timelines [2][5]. Group 2: Chip Development Timeline - The Braga chip's design was completed six months late, raising concerns about the competitiveness of future chips against Nvidia [5]. - A new intermediate chip, Maia 280, is being considered for release in 2027, which will be based on the Braga design and consist of multiple Braga chips working together [5][6]. - The Maia 400 chip, initially known as Braga-R, is now expected to enter mass production in 2028, featuring advanced integration technologies for improved performance [6][7]. Group 3: Impact on Partners - The revised roadmap has negatively impacted Marvell, a chip design company involved in the Braga-R project, leading to a decline in its stock price due to project delays and economic factors [9]. - Not all of Microsoft's chip projects are facing issues; CPU projects, which are less complex than AI chips, are progressing well [9][10]. - Microsoft's Cobalt CPU chip, released in 2024, is already generating revenue and is being used internally and by Azure cloud customers [10].

Core Insights - Microsoft is adjusting its ambitious AI chip development strategy due to delays, shifting towards a more pragmatic and iterative design approach to remain competitive with Nvidia in the coming years [1][4] - The release of the Maia 200 chip has been postponed from 2025 to 2026, with plans to launch less aggressive designs by 2028 [1][4] - Microsoft aims to reduce its dependency on Nvidia's chip procurement, which costs the company billions annually [1] Group 1: Strategic Adjustments - The delays in the development of Microsoft's second and third-generation AI chips have prompted a strategic overhaul [4] - The Braga chip's design was completed six months later than planned, raising concerns about the competitiveness of future chips against Nvidia [4] - Microsoft is considering an intermediate chip, Maia 280, to be released in 2027, which will be based on the Braga design [4][5] Group 2: Future Chip Plans - The chip initially known as Braga-R will now be called Maia 400, expected to enter mass production in 2028 with advanced integration technology [5] - The release of the third-generation AI chip, Clea, has been delayed until after 2028, with uncertain prospects [5] Group 3: Impact on Partners - The revised roadmap negatively affects Marvell, which was involved in the Braga-R project, leading to a decline in its stock price [6] - Marvell had anticipated earlier revenue from Microsoft, but delays and economic factors have impacted its performance [6] Group 4: Other Projects - Not all of Microsoft's chip projects are facing issues; the CPU project, Cobalt, is progressing well and has already generated revenue [8] - The next generation of Cobalt, Kingsgate, has completed its design and will utilize chiplet architecture and faster memory [8]

Summary of Key Points from the Earnings Call Company and Industry Overview - The discussion primarily revolves around **Microsoft** and its **Azure** cloud services, as well as the broader **data center** and **GPU** markets. Core Insights and Arguments 1. **Data Center Strategy and Demand** - Microsoft has withdrawn from certain data center commitments in Malaysia, Jakarta, and Europe, reducing capacity by 12% (2 Gigawatts) [1] - Despite this, there is strong demand for data centers in the Middle East and specific U.S. regions like Austin and San Antonio [1] - Microsoft has idled three facilities in Atlanta and exited the Stargate project, indicating a strategic shift in data center operations [1] 2. **Azure Performance and Growth Drivers** - Azure's performance exceeded expectations, driven by strong demand in general-purpose computing and big data analytics, rather than AI alone [2][3] - Major customers for Azure include TikTok and OpenAI, with GPU-as-a-service rentals contributing significantly to earnings [2] 3. **AI Revenue Breakdown** - The AI segment is projected to generate approximately $12 billion from direct GPU-as-a-service and $8 billion from AI enhancements in security and enterprise applications [3][6] - OpenAI is the largest customer for GPU services, contributing around $4.7 to $5.2 billion [6] 4. **Non-AI Growth Sustainability** - The baseline growth rate for general-purpose computing is expected to be 5% to 6% annually, with recent double-digit growth driven by external factors like tariffs [4] - The demand for data processing and analytics remains strong as companies seek to optimize costs amid supply chain challenges [4] 5. **Workforce Reorganization** - Microsoft has laid off approximately 6,000 employees and is outsourcing non-AI roles to managed service providers (MSPs) to reduce costs [5] 6. **GPU Utilization and Purchase Plans** - Microsoft has ordered approximately 1.25 million Nvidia GPUs for 2025, with a focus on Blackwell and Hopper models [24][25] - Current GPU utilization rates are high, with Blackwell GPUs prioritized for training [20][22] 7. **Capex Outlook** - Microsoft has reduced its 2025 capex from about $88 billion to $80 billion, with further reductions expected in 2026 due to delays in the Rubin program [18][19] - The percentage of capex allocated to new facilities is expected to decrease from 45-50% to 38-40% [18] 8. **Competitive Positioning** - Microsoft faces competition from AWS and GCP, with Azure focusing on high-quality customer service for large enterprises [7] - The multi-cloud strategy among clients complicates Azure's ability to attract new customers compared to AWS, which has a more direct approach with startups [7] 9. **Supply Chain and Production Issues** - There are no current shortages of GPUs, with previous issues attributed to yield and quality problems rather than demand [9][10] - The GB200 requires a redesign of data centers for deployment, indicating ongoing infrastructure adjustments [12][13] 10. **Vendor Changes and Future Plans** - Microsoft is considering switching from Marvell to Broadcom for ASIC design due to performance issues with Marvell [32] - The timeline for the Maia 300 project is set for high volume in 2027 and 2028, with a commitment to 300k units [33][34] Other Important Insights - The private sector remains free to use Chinese AI models despite government restrictions, indicating potential revenue implications for Microsoft [8] - Utilization rates are currently high but are not sustainable long-term, necessitating additional GPU purchases to maintain service levels [22] - AMD's market share is projected to be around 8% overall, while Nvidia is expected to dominate with approximately 92% [31]

Core Insights - Nvidia holds a significant share in AI training and inference markets, but competition from hyperscale computing companies developing their own XPU raises questions about sustainability [1] - Broadcom and Marvell are positioned to benefit from the demand for custom CPUs and XPUs, collaborating with major cloud providers like AWS, Google, Meta, and Microsoft [2][3] - The cost-effectiveness of these custom solutions must be significantly lower than existing offerings from Intel, AMD, Nvidia, and AMD to be viable [3] Financial Performance - Broadcom reported Q1 FY2025 sales of $14.92 billion, a 24.7% increase year-over-year, with profits reaching $5.5 billion, up 4.2 times from the previous year [5] - Marvell's Q4 FY2025 sales were $1.82 billion, a 19.9% quarter-over-quarter increase, with a net income of $200 million, marking a significant turnaround from previous losses [16] AI Revenue Growth - Broadcom's AI chip sales reached $4.12 billion in Q1 FY2025, a 77% year-over-year increase, while other semiconductor sales declined by 19.2% [11] - Marvell's AI revenue for FY2025 is projected to be around $1.85 billion, with expectations to exceed $3 billion in FY2026, driven by custom AI XPU and optical products [18][20] Market Dynamics - The IT industry is characterized by demanding clients seeking high service levels at low costs, which influences the pricing and development of custom CPUs and XPUs [3] - Broadcom's AI business is comparable in scale to Marvell's entire business, but Marvell's data center segment is rapidly growing [3][5] Future Outlook - Broadcom anticipates stable revenue of $14.9 billion for Q2 FY2025, with a projected 19.3% year-over-year growth [14] - Marvell's success in securing new hyperscale clients and developing shared AI XPU designs will be crucial for future revenue growth [20]