Author: Xu Chao, Wall Street Insights

NVIDIA CEO Jensen Huang announced the official beginning of the era of Artificial General Intelligence (AGI) and attributed this historic milestone to OpenAI's latest flagship model, Astra—this model was trained on NVIDIA chips. This statement once again places NVIDIA in a dominant position in AI infrastructure and suggests that a larger scale of computing power expansion is imminent.

On Sunday, Huang took to the X platform to congratulate the OpenAI team, stating that "AGI has arrived," and revealed that Astra was trained on over 100,000 NVIDIA Grace Blackwell NVLink72 chips. He also indicated that "400K GPUs are about to come online," sending a strong signal that NVIDIA’s computing power supply will further surge.

This statement has direct significance for the market. NVIDIA's data center business has become its core growth engine, and Huang’s forecast of adding 400,000 GPUs indicates that order demand from leading AI clients continues to expand.

OpenAI released Astra last Thursday, positioning it as "the smartest and most aligned with human values model in the world," claiming that this model can "perform the most demanding professional tasks at unparalleled speed, precision, and judgment." OpenAI President Greg Brockman told the media at the launch event, "Welcome to the era of AGI."

Brockman further predicted that in the future, when people look back at history, they will believe AGI was born "around this time, and I think this model could be it." "For my part, I really believe we have arrived," he said. Astra will begin rolling out to users this week.

OpenAI defines AGI as "highly autonomous systems that outperform humans in most economically valuable work" and positions Astra as a significant research breakthrough, marking a fundamental shift in the boundaries of tasks that can be entrusted to AI.

However, not everyone agrees with this declaration. Prominent AI researcher and critic Gary Marcus pointed out that Huang is "too hasty."

"Huang provided neither evidence nor a definition, which to me feels like trying to forcefully occupy a scientific question with corporate will," Marcus wrote on his Substack, "Declaring victory without a definition only adds more confusion to the issue."

Marcus then listed his own 10 criteria for AGI, indicating that Astra meets only one or two of them. "Measured by common definitions, Astra still falls short," he said.

Even OpenAI's own CEO Sam Altman has a fairly cautious attitude toward this concept. He recently stated on the "Sources" podcast that AGI is at best "an extremely vague term. I would almost say it’s more of a meaningless marketing term."

Regardless of how the definition war of AGI unfolds, NVIDIA's status as a core computing power supplier for leading AI companies is undisputed. In a financing announcement in March this year, OpenAI referred to NVIDIA as "the cornerstone of our infrastructure," stating, "Our training clusters and a majority of our inference stack continue to run on NVIDIA GPUs."

Leading AI companies such as Meta, Anthropic, and Google also heavily rely on high-end chips produced by NVIDIA. This demand has translated into NVIDIA's impressive financial performance—the company reported a quarterly revenue of $96.2 billion in August this year, more than doubling year-on-year; among them, data center business revenue reached $89 billion, which is the main growth source including AI chips.

Huang's post on X stated that the next batch to go online will be 400,000 GPUs, further reinforcing expectations of NVIDIA's momentum in computing power expansion. From ChatGPT to o1, and now to Astra, the technological leap over four years has been built upon NVIDIA's continuously upgraded chip foundation.

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。