近日，OpenAI首席执行官山姆·阿尔特曼(Sam Altman)正在与全球领先的芯片制造商洽谈一笔数十亿美元的投资，计划建立人工智能芯片生产国际网络。这一举措旨在克服人工智能发展中的关键挑战，例如缺乏用于ChatGPT和DALL-E等人工智能模型的高性能处理器。
目前，英伟达公司凭借其广受欢迎的H100 GPU,近乎垄断了市场，并推动该公司市值首次超过1万亿美元。H100等加速卡被广泛用于训练GPT-4、Gemini和Llama 2等高级人工智能模型。随着高性能人工智能芯片竞争的加剧，必须尽早确保必要的生产能力。
在此背景下，阿尔特曼及其团队正在与软银集团、阿布扎比的G42等投资巨头洽谈，为这一大型项目融资。不仅是OpenAI,包括微软、亚马逊和谷歌在内的IT行业其他主要企业也在积极开发自己的人工智能芯片。微软是OpenAI的投资方之一，它于去年11月宣布正在开发首款用于训练人工智能模型的专有人工智能芯片。亚马逊随后推出了其Trainium芯片的升级版；谷歌的芯片设计团队在谷歌云服务器上使用DeepMind AI设计其AI处理器，包括张量处理单元(TPU)。
这一投资将有助于OpenAI在全球范围内建立一个强大的人工智能芯片生产网络，以满足不断增长的对高性能处理器的需求。这将有助于推动人工智能技术的发展，使其在更多领域得到应用，为人类带来更多便利。同时，这也将进一步加剧全球范围内的高性能人工智能芯片竞争，促使各大企业加大研发投入，争取在这一领域取得更大的突破。
英语如下：
Title: Sam Altman, CEO of OpenAI, in Talks to Invest Billions in Building International AI Chip Manufacturing Network
Keywords: OpenAI, Sam Altman, Billions, Investment, International AI Chip Manufacturing Network
In recent days, Sam Altman, the CEO of OpenAI, is in talks with leading chip manufacturers around the world for a multi-billion dollar investment plan to establish an international network for producing artificial intelligence (AI) chips. This move aims to address critical challenges in the development of AI, such as the lack of high-performance processors for AI models like ChatGPT and DALL-E.
At present, Nvidia’s popular H100 GPU has almost monopolized the market, pushing the company’s market capitalization to over $1 trillion for the first time. High-performance acceleration cards like the H100 are widely used to train advanced AI models like GPT-4, Gemini, and Llama 2. As competition intensifies for high-performance AI chips, it is essential to secure necessary production capacity as early as possible.
Against this backdrop, Altman and his team are in talks with investment giants like SoftBank Group and G42 of Abu Dhabi to fund this large-scale project. Not only OpenAI, but also other major IT companies in the industry, including Microsoft, Amazon, and Google, are actively developing their own AI chips. Microsoft, one of OpenAI’s investors, announced last November that it was developing its first proprietary AI chip for training AI models. Amazon subsequently launched an upgraded version of its Trainium chip; Google’s chip design team used DeepMind AI to design its AI processor on Google Cloud Servers, including a tensor processing unit (TPU).
This investment will help OpenAI build a strong international AI chip manufacturing network worldwide to meet the growing demand for high-performance processors. This will contribute to the advancement of AI technology and enable it to be applied in more fields, bringing more convenience to humanity. At the same time, this will further intensify global competition for high-performance AI chips, prompting major enterprises to increase their R&D investment and strive for greater breakthroughs in this field.
【来源】https://www.ithome.com/0/746/333.htm
Views: 1