华尔街见闻·2025-11-03 11:01Core Insights - A significant infrastructure race driven by artificial intelligence is unfolding in the United States, with planned large data center projects exceeding 45 GW and attracting over $2.5 trillion in investments [1][3]. Group 1: Major Players and Projects - The expansion is primarily driven by major companies such as OpenAI, Amazon, Meta, Microsoft, and xAI, which are rapidly planning and constructing computing clusters to support increasingly complex AI models [1][3]. - Key projects include OpenAI's Stargate (1.2-1.6 GW), Frontier (1.4 GW), Lighthouse (1 GW), Project Jupiter (1.5 GW), and additional projects in Ohio and Texas, with Stargate alone representing a commitment of over $400 billion for 7 GW of capacity [4][5]. Group 2: Power Supply Challenges - The surge in power demand is creating unprecedented challenges for the U.S. electrical infrastructure, leading to a "power wall" scenario where existing grid capacity is insufficient [1][5]. - Companies are increasingly adopting a "Bring-Your-Own-Power" strategy, with many opting to build on-site power generation facilities to ensure reliable electricity supply and expedite project timelines [1][6]. Group 3: Investment and Financing Structures - The construction costs for data centers have escalated, with costs exceeding $1,700 million per MW, and OpenAI's Stargate project reflecting a staggering $5,700 million per MW when including IT equipment [4][8]. - Private equity firms and specialized infrastructure funds are playing a crucial role in financing these projects, exemplified by Blue Owl Capital's $15 billion joint venture with Crusoe for the Stargate 1 project [8]. Group 4: Energy as a Service (EaaS) Model - The rise of the "Energy as a Service" (EaaS) model is evident, with energy companies like Williams entering long-term power purchase agreements with data center operators, investing billions in dedicated power generation facilities [9]. Group 5: Supply Chain and Labor Challenges - The explosive demand is straining the power equipment supply chain, with heavy gas turbine prices rising by 50% in less than two years and extended delivery times [10][11]. - Equipment manufacturers are facing challenges related to component shortages and labor, prompting some companies to acquire second-hand or unused equipment to meet their needs [11].