NVIDIA Slams AI Bubble Talk, Market Talks Back

NVIDIA shares popped immediately after a stellar earnings report last night, including a raise in its market guidance for 2026. But it didn't last long: By noon ET today shares were in the red, showing the market is growing increasingly uneasy about the risk posed by NVIDIA's outsized influence on the AI economy, particularly regarding its deals with OpenAI.

NVIDIA exceeded Wall St. expectations in posting revenues of $57 billion, up 62% year-over-year (y/y) for the third quarter of its fiscal year 2026. Of that, datacenter revenue accounted for $51.2 billion, up 55% y/y, thanks to Blackwell adoption and a surge in sales of Spectrum-X networking gear (which rose 162%). Investors seemed impressed with non-GAAP gross margin of 73.6% and adjusted net income of $32 billion. More importantly, its guidance for $65 billion in sales next quarter and non-GAAP gross margin of 75% blew away the expectations of most Wall St. analysts.

Further, the company anticipates more than $500 billion in business through the next few quarters. “Currently, we have visibility to a half a trillion dollars in Blackwell and Rubin revenue from the start of this year through the end of calendar year 2026,” said CFO Colette Kress on the earnings call. And later she said: “[T]here’s definitely an opportunity for us to have more on top of the $500 billion that we announced.”

Today's adverse market reaction to a seemingly brilliant earnings report shows that more might be happening below the surface of the AI infrastructure market. If NVIDIA can blow away its numbers but the stock falls, what kind of positive surprise is left that might draw new investors?

Let's dive in.

NVIDIA’s Management Cheerleaders

Throughout the call, Kress and CEO Jensen Huang touted the strength of NVIDIA’s sales and backlog, its technological leadership, and the reliability of its supply chain. “There’s been a lot of talk about an AI bubble,” Huang said. “From our vantage point, we see something very different.”

NVIDIA CEO Jensen Huang

To prove that point, Kress and Huang cited a market still in transition but driven by growth in accelerated computing, the introduction of ever-more-powerful AI models, and agentic AI applications. Through a series of examples, management led listeners through a convincing profile of a dominant player in a market NVIDIA predicts will hit $3 trillion to $4 trillion by the end of the decade.

Some investors seemed cheered by the results. “All in all, we view this as a thesis-confirming quarter, highlighting Nvidia’s sustained position as the leading provider of accelerated computing infrastructure,” wrote William Blair analyst Sebastien Naji in a note following the call.

Questions and Circular Answers

But not everyone is convinced that NVIDIA’s earnings put the bubble talk to bed. “Anyone who thinks that NVIDIA’s extraordinary quarterly earnings release is proof that there is no AI bubble has no understanding of how AI works and the circular relationships between the key players,” wrote Michael Hutchens, founder at financial modeling and reporting software firm Modano, in a LinkedIn post.

There is concern about the role of OpenAI in the AI market, particularly about that company’s role in creating a circular pattern of spending in which NVIDIA (via $100 billion), Oracle, neocloud CoreWeave, and others are funneling billions (in borrowed money in the cases of Oracle and CoreWeave) into datacenter buildouts on pledges of infrastructure usage by OpenAI. Yet OpenAI is reportedly on track to burn an estimated $8.5 billion in cash this year despite expecting to score $13 billion in annual sales. And OpenAI doesn’t anticipate profitability until 2029 or 2030.

Not to worry, said Jensen Huang on the earnings call, circling back to his aggressive optimism. “Don’t just look at what is in the press....,” he said. “So we invest in OpenAI for a deep partnership and co-development to expand our ecosystem and to support their growth. And, of course, rather than giving up a share of our company, we get a share of their company…. And so I fully expect that investment to translate to extraordinary returns.”

Others agree that if OpenAI continues to build out good AI models, its future should be positive. Yet ChatGPT 5’s release in August 2025 didn’t go smoothly, and a new release GPT 5.1 launched in November to relieve the cacophony of complaints. There seem to be concerns in some quarters that the recent ChatGPT hiccup hints at a slowdown in OpenAI innovation. All of which makes OpenAI’s future unpredictable and fuels those niggling worries about a bubble.

There are a few wild cards, too: It's not clear how obstacles such as power constraints and supply chain bottlenecks for components such as memory chips could affect NVIDIA's growth. Asked about this last night, CEO Huang said:

"[W]e're incredibly good at managing our supply chain. We have great partners that we've worked with for thirty-three years. And so the supply chain part of it, we're quite confident.... We've established a whole lot of partners and so we have a lot of routes to market. And very... importantly, our architecture has to deliver the best value to the customers that we have."

Then there's the matter of NVIDIA's China situation—specifically, will there ever be a time when NVIDIA can once again sell chips to PRC customers, even if those customers agree to buy them? On this point, Kress said NVIDIA is intent on maintaining engagement with both the U.S. and China governments and continues to see the importance of including China in international sales.

Futuriom Take: While NVIDIA’s latest earnings report was impressive, there remain questions about the future of AI demand, particularly for datacenter infrastructure funded in part by NVIDIA for use by OpenAI. And the market's reaction to the earnings shows that NVIDIA's ebullient cheerleading may be somewhat counter-productive.