智通财经网·2025-08-14 08:48Core Viewpoint - The controversial proposal by President Trump to impose a revenue-sharing scheme on AI chip sales from Nvidia and AMD to China is raising market concerns about potential erosion of U.S. corporate profits and innovation advantages [1] Group 1: Market Reactions - Investors are worried that the revenue-sharing model may extend beyond Nvidia and AMD, potentially affecting more semiconductor leaders and their fundamentals [1] - Analysts express skepticism about the precedent set by the U.S. government in imposing penalties on chip exports, yet acknowledge that receiving 85% of revenue is better than receiving nothing [2] Group 2: Technical Performance - Nvidia's H20 AI chip, approved for export to China, has significantly reduced performance compared to its predecessor H100, particularly in AI training tasks [2] - However, the H20 chip has been optimized for AI inference, showing competitive performance in certain scenarios, which may provide Nvidia with a unique advantage in the Chinese market [2] Group 3: Government Revenue Initiatives - The Trump administration's revenue-sharing agreement with Nvidia and AMD is seen as a potential model for similar initiatives across other industries [4] - The U.S. government is exploring various revenue sources, including the sale of "gold cards" for immigration and "golden shares" in companies, to bolster its fiscal position [6] Group 4: Legislative Concerns - The move has raised eyebrows among hardliners in Congress, who question the legal basis of such revenue-sharing agreements and their implications for national security [6] - Concerns are growing that this could set a precedent for further government fees on U.S. companies engaged in international trade [3] Group 5: Future Implications - The long-term impact of these revenue-sharing measures on U.S. corporate profits and innovation remains uncertain, with some experts warning of potential negative consequences [7] - The Trump administration argues that these measures could enhance U.S. influence and market share in the AI sector, although skepticism persists regarding their effectiveness [7]