Nvidia CEO claims AGI arrival, but markets remain skeptical

Nvidia chief Jensen Huang posted on X that OpenAI's latest model, Astra, meets the criteria for artificial general intelligence (AGI) after being trained on tens of thousands of Grace Blackwell chips. The claim drew mixed reactions, with some commentators praising Nvidia's position and others questioning Huang's motives.

When markets opened after the Labor Day weekend, shares of companies tied to OpenAI such as CoreWeave and SoftBank rose, while Nvidia's stock slipped 2 percent. Analysts noted that the firms most exposed to OpenAI's technology were the ones that benefited, whereas Nvidia, despite posting record quarterly revenue of $96 billion, saw little movement.

"If you look in a slightly different place, you’ll see that the market is responding," said Gil Luria, head of technology research at D.A. Davidson. He added that Astra places OpenAI back at the forefront of model development, but warned that Nvidia's size makes further growth hard to sustain.

Critics of the AGI label pointed to academic definitions that require an AI to match the cognitive versatility of a well-educated adult. Economist Basil Halperin said, "We just absolutely have not achieved AGI, even though the models are astounding," citing a personal test where a GPT-5.6 model required extensive supervision to complete a simple task.

Halperin argued that true evidence of AGI would appear in macroeconomic indicators such as real interest rates, which tend to rise with higher growth expectations. He noted that real rates have increased since 2021 but not to the levels that would signal a transformative AI impact, and that past model releases have been associated with modest declines in long-term Treasury yields.

Overall, analysts suggest that while Astra represents a technical advance, investors remain cautious, viewing the claim of AGI as speculative rather than a driver of immediate market value.