Nvidia CEO Jensen Huang Claims AGI Has Arrived as OpenAI Launches New Model

Earlier this week, Jensen Huang, the Nvidia (NVDA) cofounder and CEO, posted something bold on social media. The comment landed just as OpenAI rolled out a brand-new model that the company says beats everything it has built before.

For anyone tracking Nvidia stock, the timing was hard to miss.

Jensen Huang Says the AGI Era is Here

According to Fox Business:

• OpenAI unveiled its newest model, called GPT -6 Astra, last Thursday, bringing major upgrades to computer use, coding, science, and everyday professional work.
• Astra began rolling out to a limited group of enterprise customers. It is also becoming available to ChatGPT Plus, Pro, Business, and Enterprise users, as well as through the OpenAI API and Amazon Web Services. 

Huang wrote a post on X congratulating the OpenAI team on achieving artificial general intelligence (AGI). He noted that Astra was trained using more than 100,000 Nvidia Grace Blackwell NVLink72 chips. Huang pointed out that in just four years, the industry moved from the original ChatGPT to the reasoning model o1 to Astra. Huang also added that another 400,000 GPUs are coming online next.

The tech industry has no single agreed-upon definition of AGI. Generally, it refers to a system with human-level or better reasoning that can pick up new skills without retraining from scratch.

OpenAI defines AGI as highly independent systems that outperform people at most work that has real economic value, Fox Business noted. Nvidia and Huang have used a different yardstick, describing AGI as an AI system able to pass professional certification exams and standardized tests with scores near the very top.

OpenAI President Greg Brockman told reporters ahead of the launch that Astra represents a turning point, saying it is not unreasonable to feel the industry has entered the AGI era.

What GPT 6 Astra Can Do

OpenAI calls Astra its most intelligent and best-aligned model to date. The company said Astra sets new highs on tough benchmarks, including a math test called FrontierMath Tier 4, where it scored 98%, posted near-perfect results on ARC AGI 3, and a cybersecurity test called ExploitBench.

Beyond test scores, OpenAI said Astra can handle repetitive computer work such as filling out online forms, updating customer records, and organizing calendars. It can also research topics, draft documents, build spreadsheets and presentations, and even test software it just built. 

OpenAI CEO Sam Altman told Fox Business that one of the most impressive traits was Astra flagging issues users had not even thought to ask about. He gave an example of researching a complex chip supply chain, where the model returned gaps in the original request the user never mentioned.

None of this happens without massive computing power, and that is where Nvidia comes in. On its most recent earnings call, Nvidia reported second-quarter revenue of $96 billion, more than double what it posted a year earlier, with data center revenue reaching $89 billion. The company guided for $108 billion in the current quarter and said it expects fiscal 2028 revenue to grow roughly 70% year over year, amid supply constraints. 

Huang told analysts on that call that AI agents burn through 15 to 100 times more compute than a person doing the same task, since agents plan, reason, and use tools across multiple steps. He said Nvidia's growth is not limited by technology or customer demand, but by how much compute the company can actually supply.

Huang also touched on AGI himself during the call, saying that for many tasks, in a lot of ways, "we could say that we have already achieved AGI." He added that chasing a milestone matters less than the fact that AI is now doing useful, profitable work at scale.

What's Next for Nvidia's Stock Price

Nvidia's next-generation Vera Rubin platform is already shipping, and the company expects it to be its fastest product ramp ever. Huang said demand from hyperscalers, AI labs, sovereign nations, and enterprises keeps climbing faster than supply chains can keep pace.

With OpenAI now leaning on Nvidia hardware to power Astra, and rivals racing to keep up, investors will likely keep watching Nvidia's stock price closely heading into 2027 as the AGI debate plays out in public.

Out of the 50 analysts covering Nvidia stock, 45 recommend “Strong Buy”, three recommend “Moderate Buy”, one recommends “Hold”, and one recommends “Strong Sell”. The average Nvidia stock price target is $325, above the current price of $226. 

On the date of publication, Aditya Raghunath did not have (either directly or indirectly) positions in any of the securities mentioned in this article. All information and data in this article is solely for informational purposes. For more information please view the Barchart Disclosure Policy here.