AI端侧深度报告之AI PC：PC助力端侧AI规模化拓展，算力、存储、能耗升级显著

Investment Rating - The report recommends a "Buy" rating for AI PC-related companies, highlighting their potential to drive a new wave of innovation in the PC market [4]. Core Insights - The trend of cloud AI expanding to edge AI is evident, with AI PCs facilitating the scaling of hybrid AI. Cloud AI focuses on high performance and density, while edge AI emphasizes low power consumption and high efficiency. The collaboration between cloud and edge AI for computational load is expected to become a norm [3][19]. - AI PCs are projected to significantly boost the PC market, with global shipments expected to reach 51 million units in 2024 and 208 million units by 2028, reflecting a compound annual growth rate of 42% from 2024 to 2028 [3][4]. Summary by Sections AI PC Development - AI PCs are designed to support low-power, high-efficiency applications, making them ideal for hybrid AI deployment. They are expected to lead to a significant upgrade in the PC market [2][3]. - The necessary specifications for AI PCs include a minimum of 40 Tops of NPU performance, with current models from Qualcomm exceeding this requirement, while Intel and AMD are expected to release upgraded models in 2024 [3][32][37]. Market Projections - Canalys forecasts that the penetration rate of AI PCs will increase to 19% in 2024, 43% in 2025, and 55% in 2026, indicating a rapid adoption of AI technology in personal computing [3][4]. Investment Recommendations - The report suggests investing in companies such as Chipsea Technology, Huichuangda, and others, which are positioned to benefit from the anticipated growth in the AI PC market [4]. Hardware and Software Ecosystem - AI PCs require significant hardware upgrades, including at least 16GB of memory for basic AI models, with higher specifications for advanced models. The report emphasizes the need for improved SSD performance and capacity [48]. - The software ecosystem for AI PCs is still developing, with companies like Microsoft introducing platforms to enhance AI application development and deployment [49][51].