Investment Rating - The report maintains a "Buy" rating for Advanced Micro Devices (AMD) with a target price of $340 [6]. Core Insights - AMD has entered a strategic multi-year partnership with Meta, deploying up to 6GW of AMD Instinct GPUs, which has positively impacted AMD's stock price, increasing by approximately 8.8% to $213.84 [1]. - The collaboration with Meta is similar to AMD's previous partnership with OpenAI, utilizing a dual mechanism of "compute + equity" [2]. - AMD's AI business visibility is expected to improve significantly, with projected Non-GAAP EPS potentially exceeding $20 in the next 3-5 years [1]. Summary by Sections Strategic Partnerships - AMD's partnership with Meta includes the deployment of 1GW of MI450 GPUs and the sixth-generation EPYC (Venice) CPUs, with shipments expected to start in the second half of 2026 [1]. - The partnership also involves performance-based warrants for up to 160 million shares, linked to GPU shipment milestones and stock price thresholds [1]. Revenue Projections - AMD anticipates significant revenue from AI orders, with an estimated 2.2GW deployment in 2026 from OpenAI, Meta, and Oracle, translating to approximately $13.75 billion in AI revenue [2]. - The revenue estimates for 2026-2028 have been revised upward by 15%/20%/17% to $65.8 billion, $88.4 billion, and $102.9 billion, respectively [4]. Financial Metrics - Non-GAAP net profit estimates for 2026-2028 have been increased by 13%/20%/16% to $15.8 billion, $20.8 billion, and $23.7 billion, respectively [4]. - The report highlights AMD's AI chips gaining recognition, suggesting a potential alignment of its PE ratio with Nvidia's historical averages [4][14]. Market Positioning - Meta's accelerated investment in AI infrastructure, with a capital expenditure (Capex) guidance of $115-135 billion for 2026, indicates a robust demand for AMD's products [3]. - AMD's AI chip roadmap is expected to compete directly with Nvidia's offerings, enhancing its market position [14].

今日早参 2026 年 2 月 26 日 张继强 研究所所长、固收首席研究员 座机：13910012450 邮箱：zhangjiqiang@htsc.com 陈慎 房地产行业首席研究员 座机：021 38476038 邮箱：chenshen@htsc.com 今日热点 风险提示：房地产政策波动，房地产市场复苏不及预期，部分房企经营风 险。 研报发布日期：2026-02-25 研究员 刘璐 SAC:S0570519070001 SFC:BRD825 固定收益：AI 叙事的转变 ——基本面观察 2 月第 2 期 2026 年以来，全球 AI 叙事正在经历一次重要的边际变化，我们至少看到了 三层叙事的转变。 第一层叙事：对 Scaling Law 开始出现一些分歧。 过去几年 AI 投资的核心引擎来自于 Scaling Law 的经验规律：模型越大、数 据越多、算力越强，性能越好。但这条规律正在出现一些裂痕。 风险提示：AI 开支不及预期，地缘政治风险。 研报发布日期：2026-02-25 研究员 张继强 SAC:S0570518110002 SFC:AMB145 吴靖 SAC:S0570523070006 房地产 ...

Summary of AMD Conference Call Company Overview - The conference call discusses AMD's strategic initiatives and market outlook, particularly in the context of AI and data center opportunities. Key Points Industry Dynamics - Major companies are significantly increasing capital expenditures supported by free cash flow, benefiting the entire ecosystem [2][3] - Confidence in AI's commercial prospects is rising, with real application cases emerging, leading to productivity improvements and reduced inference costs [2][3] - The data center market presents a trillion-dollar opportunity, with accelerators capturing a significant share, including general-purpose and custom ASIC chips [2][4] AMD's Strategic Partnerships - AMD signed a multi-year 6 GW collaboration agreement with OpenAI, which is a defensive agreement based on performance warrants [2][5] - The first gigawatt-level product is planned for deployment in the second half of 2026, with gradual scaling in 2027 [2][5] - Collaboration with Meta on the Helios rack, designed to meet OCP standards, aims for high-density deployment [2][6] Product Focus and Market Position - AMD does not sell heavy rack-level systems but focuses on high-value components like GPUs, CPUs, and networking [2][7] - The client business has seen a 60% revenue growth over three consecutive quarters, driven by improved average selling prices (ASP) and a shift towards high-end PC and enterprise markets [3][12] - The company anticipates that GPU gross margins will gradually rise as business scales and solutions optimize [2][7] Market Opportunities and Challenges - The data center market is expected to see a shift from general silicon chips to custom silicon chips, with ASICs projected to capture 20%-25% of the market [4][5] - AMD's addressable market is estimated at one trillion, focusing on data center opportunities, including accelerators and CPUs, while competitors may include broader infrastructure components [4][5] - The company is closely monitoring the Chinese market dynamics and has received some licenses to operate within regulatory frameworks [10][11] Future Outlook - AMD aims for overall gross margins to reach 55%-58% in the long term, driven by investments in high-margin business areas and expanding market share [7][12] - The company is focused on reducing total cost of ownership (TCO) in data centers while supporting open standards, with significant developments in their Infinity Fabric technology [9][10] Competitive Landscape - AMD is evaluating customer AI workload characteristics and maintaining flexibility in hardware design, with ongoing assessments of general versus dedicated silicon needs [8][9] - The company is positioned well in the server market, with a rapidly expanding enterprise customer base and significant CPU demand driven by inference computing [12] Conclusion - AMD is strategically positioned to capitalize on the growing AI and data center markets, with strong partnerships, a focus on high-value products, and a commitment to innovation and flexibility in its offerings [2][3][4][5][6][7][8][9][10][11][12]

Summary of AMD Conference Call Company Overview - **Company**: AMD (Advanced Micro Devices) - **Industry**: Semiconductor, focusing on high-performance computing and AI Key Points and Arguments 1. **Strategic Shift**: AMD has strategically shifted its R&D focus towards high-performance computing and AI, resulting in data center business growth exceeding 60%, significantly above market averages, indicating sustained demand for computing power [2][4][9] 2. **Unique Technology Portfolio**: AMD is the only semiconductor company with CPU, GPU, and FPGA technologies, having pioneered large-scale production of triad technology, which enhances its competitive edge in the data center market [2][5] 3. **GPU Market Dominance**: Despite the value of ASICs in specific scenarios, AMD believes GPUs will dominate the accelerator market over the next five years due to their programmability and flexibility, with ASIC market share expected to be between 20%-25% [2][7] 4. **Customized Solutions**: AMD collaborates deeply with clients through semi-custom business models, leveraging its IP and client expertise to optimize specific workloads, rather than merely fulfilling client requests [2][8] 5. **Super Cycle in Computing**: AMD perceives the current phase as the early stage of a decade-long super cycle driven by computing power, with AI investments expected to enhance productivity and product optimization [2][9] 6. **Full-Stack Solutions**: Through acquisitions like ZT Systems and partnerships with Sanmina, AMD aims to provide full-stack solutions that reduce the time clients need to establish complex infrastructures [2][11] 7. **Collaboration with OpenAI**: AMD has partnered with OpenAI to optimize hardware and software design, which is expected to generate significant revenue, enhancing market recognition of AMD's technological capabilities [3][13] 8. **Client Diversity**: AMD operates as a general supplier, maintaining long-term relationships with major hyperscale computing companies, which mitigates concerns regarding customer concentration [14] 9. **CPU Market Trends**: Recent months have seen a significant increase in CPU demand, driven by a major update cycle and delayed AI-related capital expenditures catching up, suggesting substantial growth in the CPU market over the next four to five years [15][16] 10. **Future Market Share**: AMD is well-positioned to continue gaining market share in the data center sector, with the upcoming Venue CPU expected to further solidify its leadership [17] 11. **Client PC Market Ambitions**: AMD aims for ambitious market share growth in the client PC sector, currently holding a market share in the mid-high 20% range, with strong performance in the desktop gaming and high-end laptop markets [18] 12. **Memory Price Impact**: While rising memory prices may have minor effects, they are not expected to significantly impact overall market demand [19] 13. **Growth Bottlenecks**: Key factors that could limit growth include advanced technology access, high bandwidth memory, and power supply issues, particularly in data centers [20] Additional Important Insights - AMD's focus on integrating various computing units (CPU, GPU, FPGA) allows for flexibility in addressing different workloads, enhancing its innovation in AI and software investments across its product line [12] - The company emphasizes the importance of collaboration with clients to ensure successful outcomes and differentiate its offerings in a competitive landscape [14]

Core Insights - Oracle and AMD are expanding their long-term partnership to enhance AI capabilities, with Oracle's cloud infrastructure set to deploy a significant AI supercluster powered by AMD's Instinct MI450 series GPUs starting in Q3 2026 [9][10]. Group 1: AI Supercluster Details - The AI supercluster will initially deploy 50,000 Altair GPU slots, with plans for gradual expansion in 2027 and beyond [6][9]. - The MI450 GPU is expected to deliver approximately 40 petaflops of peak computing power with 432 GB of HBM4 memory and a memory bandwidth of 19.6 TB/s [2][11]. - The Helios rack design will support up to 128 MI450X GPUs, providing a total of 1.45 exaflops at FP8 precision and 2.9 exaflops at FP4 precision [5][10]. Group 2: Technical Specifications - The MI450 series will utilize TSMC's 2nm process technology, enhancing performance and efficiency [2]. - Each Helios rack will feature advanced networking capabilities, including UALink for GPU interconnectivity and Pensando DPU technology for enhanced security and performance [12][14]. - The architecture is designed to optimize performance density, cost, and energy efficiency, crucial for large-scale AI operations [12][13]. Group 3: Market Context and Future Outlook - The demand for large-scale AI capacity is accelerating, necessitating flexible and open computing solutions [10]. - Oracle's Acceleron network architecture aims to reduce equipment layers in large-scale AI deployments, indicating a strategic move towards more integrated solutions [7][11]. - The anticipated total cost for the deployment of 700 racks is estimated to be between $3.5 billion and $4 billion, reflecting the high demand and limited supply of GPUs [6][7].

Investment Rating - The report maintains an "Outperform" rating for the communication industry, indicating an expected performance exceeding the benchmark index by more than 10% [1][41]. Core Insights - AMD has launched the MI350 series GPUs and showcased plans for the MI400 series, intensifying competition among the "AI Triad" in North America [2][14]. - The data center AI accelerator market is projected to grow over 60% annually, reaching $500 billion by 2028, with inference market growth expected to outpace training [2][14]. - The report highlights the ongoing valuation recovery in the overseas computing chain, driven by improved earnings confidence for 2026 and a return of value funds [3][15]. - The GTC Paris summit emphasized the importance of sovereign AI and industrial cloud as significant growth areas for computing power in Europe [4][15]. - The report suggests focusing on companies within the computing power supply chain, particularly in edge computing and endpoint devices [5][16]. Summary by Sections Industry Dynamics - AMD's MI350 series features 185 billion transistors and 288GB HBM3e, with performance improvements over the MI300 series exceeding 100% [2][14]. - The MI350X and MI355X GPUs demonstrate up to 35 times the inference performance of the MI300X [2][14]. - The report notes that the European market is becoming a key area for AI deployment, with significant investments in sovereign AI systems [4][15]. Mergers and Acquisitions - East Mountain Precision announced plans to acquire Source Photonics for up to $629 million, indicating a trend of consolidation in the optical communication sector [5][16][17]. - Source Photonics reported a revenue of 2.93 billion and a net profit of 400 million for 2024, with a 32% year-on-year growth in Q1 revenue [7][17]. Market Performance - The overall market saw a decline during the week of June 9-13, 2025, with the communication index dropping by 0.78% [8][18]. - The optical module sector showed resilience with a weekly increase of 3.12%, indicating strong performance amidst broader market challenges [8][18].