Nvidia(US:NVDA) 硬AI·2026-03-11 09:11Core Viewpoint - AI is not merely a model or application but a foundational infrastructure akin to electricity and the internet, requiring significant investment for future development [5][6][20]. Group 1: AI Industry Structure - The AI industry is structured as a "five-layer cake" consisting of energy, chips, infrastructure, models, and applications, with strong interdependencies among these layers [7][9][26]. - The foundational layer is energy, which is essential for real-time intelligence generation, as every generated token relies on energy conversion into computational power [9][24]. - Above energy is the chip layer, which efficiently transforms energy into computational power, crucial for AI workloads that demand high parallel computing capabilities [9][24]. - The infrastructure layer includes systems like land, power delivery, cooling systems, and networks that enable thousands of processors to work together, forming "AI factories" designed to produce intelligence [9][24]. - The model layer encompasses AI models that can understand various types of information, with significant advancements occurring in fields like protein AI and robotics [10][24]. - The application layer is where economic value is created, with examples including drug discovery platforms and autonomous vehicles, all utilizing the same underlying technology stack [10][25]. Group 2: Current State and Future Needs - Currently, global investment in AI infrastructure is in the thousands of billions, but future needs are projected to require investments in the tens of trillions [6][13][27]. - The construction of chip factories, server assembly plants, and AI data centers is accelerating, potentially becoming one of the largest infrastructure projects in human history [13][27]. - There is a growing demand for skilled labor in AI data center construction, including electricians and network engineers, highlighting the need for a diverse workforce [13][27]. Group 3: Role of Open Source Models - Open source models play a critical role in the AI ecosystem, with many organizations relying on them to drive AI development [15][29]. - The advancement of open source models can stimulate demand across the entire technology stack, as seen with the example of DeepSeek-R1, which increased the need for training power and infrastructure [15][30]. Group 4: Broader Impacts of AI - AI is expected to transform not only the software industry but also sectors like energy, manufacturing, and labor structures, indicating a significant industrial-level transformation [18][31]. - The ongoing development of AI is anticipated to enhance productivity across various fields, such as radiology, where AI assists in image analysis while increasing the demand for radiologists [27][28].