Nvidia(US:NVDA) 3 6 Ke·2026-01-22 01:31Core Viewpoint - The construction of AI infrastructure is a significant global initiative, requiring substantial investment in energy, chips, and data centers, as emphasized by NVIDIA's founder Jensen Huang at Davos [1][10][11]. Group 1: Energy as the Foundation - The first priority in AI infrastructure is electricity, which must be stable, abundant, and capable of supporting high-density, low-latency demands [2][4]. - The need for energy is critical; without sufficient power, AI systems cannot operate effectively, highlighting the necessity for a comprehensive energy supply system [4][5]. - Countries aiming to build AI capabilities are first assessing their electricity supply [5][6]. Group 2: Chip and AI Factory Developments - Global investments are being made in chip and AI factories, with TSMC planning to build 20 new chip plants and companies like Quanta, Wistron, and Foxconn constructing 30 AI computer factories [7][8]. - The storage of data is equally important, with Micron investing $200 billion in memory production, alongside similar commitments from Samsung and SK Hynix [8][10]. - This construction wave is not limited to a few companies but represents a global trend in building the necessary hardware for AI [9][10]. Group 3: The Role of AI Models - AI models are just one layer in a five-layer structure, with energy, chips, and cloud services forming the foundational layers [13][14]. - The focus is shifting from merely developing models to effectively applying them in real-world scenarios, which is where true value is generated [17][18]. Group 4: Emergence of AI-native Companies - 2025 is projected to be a peak year for venture capital investment in AI-native companies, which leverage existing models for practical applications in various industries [19][20]. - These companies are transforming sectors like pharmaceuticals and finance by integrating AI into their processes, leading to significant operational efficiencies [21][22]. - The growth of AI-native companies necessitates an expansion of foundational infrastructure to support their needs [22]. Group 5: Workforce and National Involvement - The construction of AI infrastructure is creating high-demand jobs for skilled laborers, such as electricians and steelworkers, with salaries rising significantly [24][25]. - Contrary to fears of job displacement, AI is enhancing roles in fields like radiology and nursing by automating repetitive tasks, allowing professionals to focus on more complex responsibilities [26][28]. - Huang emphasizes the importance of participation from developing countries in AI infrastructure, suggesting that they can leverage existing models and local knowledge to engage in AI development [30][31]. Group 6: Market Dynamics and Future Outlook - The current market is characterized by shortages rather than bubbles, with rising prices for GPUs indicating strong demand for AI infrastructure [32][34]. - Investment in AI labs and infrastructure is increasing as companies recognize the necessity of robust foundational elements for AI applications [34][35].