英伟达计划投资千亿美元与OpenAI合作部署AI算力集群

Core Insights - Nvidia plans to invest $100 billion in OpenAI as part of a strategic partnership, which will enable OpenAI to deploy at least 10GW of AI computing power using millions of Nvidia GPUs [1][2] - The first GW of computing power is expected to be deployed in the second half of 2026, utilizing Nvidia's next-generation Vera Rubin platform [1] - This partnership indicates that OpenAI is moving away from its reliance on Microsoft, which has been its primary provider of computing power since 2019 [2] Group 1 - Nvidia and OpenAI have announced a strategic partnership to enhance AI computing capabilities [1] - OpenAI will utilize millions of Nvidia GPUs to establish a significant AI computing power infrastructure [1] - Nvidia's investment will be phased with each GW of computing power deployed, starting with a total commitment of $100 billion [1] Group 2 - The partnership allows OpenAI to reduce its dependency on Microsoft, which has historically provided the majority of its computing resources [2] - Microsoft has a profit-sharing agreement with OpenAI, allowing it to receive 49% of profits until it recoups its $13 billion investment [2] - OpenAI is diversifying its sources of computing power and capital by partnering with companies like SoftBank, Oracle, and Nvidia [2]