Jensen Huang Says “AGI Has Arrived” After OpenAI Releases Astra

Nvidia CEO Jensen Huang congratulated OpenAI after the release of Astra, calling it the company’s newest and most powerful AI model. In a post on X, he wrote: “From ChatGPT to o1 to Astra in 4 years. AGI has arrived. Congratulations @OpenAI team.”

Huang also noted that Astra was trained on Nvidia’s chips, tying OpenAI’s model milestone directly to the infrastructure behind frontier AI systems.

OpenAI unveiled Astra on Thursday and described it as the world’s “most intelligent and aligned model.” The company said Astra can perform “the most demanding professional work with unmatched speed, accuracy, and judgment.”

OpenAI defines AGI as “highly autonomous systems that outperform humans at most economically valuable work.” The company said Astra is rolling out to customers this week.

Not everyone accepts Huang’s conclusion. AI researcher and critic Gary Marcus said Huang had “jumped the gun,” arguing that the claim came without evidence or a clear definition.

Marcus wrote that “declaring victory without a definition simply muddies the waters” and said Astra meets only one or two points in his own 10-point definition of AGI. Sam Altman has also described AGI as “a very poorly defined term” and compared it to an “irrelevant marketing term.”

The article underscores how dependent leading AI companies have become on Nvidia’s advanced chips. OpenAI has called Nvidia “the foundation of our infrastructure” and said its training fleet and most of its inference stack continue to run on Nvidia GPUs.

Business Insider reported that Nvidia posted $96.2 billion in quarterly revenue in August, with its data center business generating $89 billion. Huang also said “400K GPUs” are coming online next, pointing to more compute capacity for training and running AI models.