美国专家来中国转了一圈：AI比赛已经结束了

Core Viewpoint - The article discusses the significant gap between China and the United States in AI development, primarily attributing this to differences in energy infrastructure and supply, suggesting that the competition may already be concluded in favor of China [1][2][20]. Group 1: Energy Infrastructure - A key argument is that energy supply is crucial for AI development, and China has effectively addressed its energy challenges, providing stable and affordable electricity [6][20]. - In contrast, the U.S. faces significant issues with its aging power grid, with 70% of transmission lines over 25 years old, making it difficult to meet modern energy demands [30][31]. - The U.S. has a low reserve capacity for electricity, around 15%, compared to China's 80% to 100%, leading to vulnerabilities during disasters and price surges [37][38]. Group 2: AI Development Landscape - Chinese AI companies are strong but struggle with profitability due to lower pricing of products and services [16]. - The U.S. tech companies are criticized for their short-sightedness, focusing on immediate profits rather than long-term infrastructure investments, which hampers AI progress [45][47]. - The article highlights that the U.S. is experiencing a significant backlog of energy projects waiting for grid connections, which has doubled since 2020 [33][36]. Group 3: Expert Insights - Rui Ma, a Chinese-American expert, emphasizes that energy supply is taken for granted in China, contrasting with ongoing debates in the U.S. about energy consumption and grid limitations [21][22]. - The article references Hinton's concerns about the short-term focus of U.S. tech companies, which he believes undermines the responsible development of AGI [50][56]. - Hinton's recent statements suggest a growing disillusionment with Silicon Valley's approach to AI, indicating a potential shift in focus towards China for responsible AI development [57][58].