At an internal meeting a week ago, Nvidia Nvidia CEO Jensen Huang admitted that despite delivering "incredible" results, market expectations for the company have reached such a high level that it's in a "no-win" situation. He bluntly stated, "If we deliver a bad quarterly report, even just a little bit off, even a little bit unstable, the whole world will collapse."
Such predictions are becoming a reality. Despite delivering financial results that exceeded market expectations, the market sentiment began to shift after Google released Gemini 3. Google's model uses its own TPU chip, rather than Nvidia GPUs, and more importantly, the industry believes it has "surpassed" OpenAI's GPT model.
On November 25th, during trading on the US stock market, Nvidia's stock price once plummeted by more than 7%, instantly wiping out nearly $350 billion in market value. Although the decline subsequently narrowed, it still closed down 2.59% at $177.82, a new closing low in more than two months. Nvidia had reached an all-time high of $212 per share on October 29th, at which time its total market capitalization reached $5.15 trillion. However, its stock price has since fallen by about 16% from that all-time high, resulting in a market capitalization loss of over $800 billion.
Nvidia issues urgent statement
On the same day, Nvidia made a rare public statement, saying that its technology remains a generation ahead of the industry and is the only one capable of running all artificial intelligence. (AI) models and platforms applicable to all computing scenarios. Analysts believe this move by Nvidia is a response to Wall Street concerns that its dominance in AI infrastructure may be threatened by Google chips.
Nvidia posted on the social media platform X: "We are delighted with Google's success and their tremendous progress in artificial intelligence , and we will continue to supply Google. Nvidia is a whole generation ahead of the industry - the only platform that can run all AI models and be deployed in all computing scenarios."
Nvidia added, "Compared to ASIC (Application-Specific Integrated Circuit) chips designed specifically for a particular AI framework or function, Nvidia offers higher performance, greater versatility, and better substitutability."
But the market trend has begun to change.
Just 27 days after Nvidia reached its peak market capitalization, investors began to question the real demand for AI computing power and whether AI investments could translate into sufficient returns. Not only did Bridgewater significantly reduce its Nvidia holdings in the third quarter, and "Silicon Valley venture capital godfather" Peter Thiel liquidate his entire Nvidia position in the same quarter, but Michael Burry, the "big short" who shorted real estate before the 2008 subprime mortgage crisis, also recently criticized the cyclical trading of tech companies around Nvidia on social media.
Michael Burry posted a chart on social media depicting the investment and procurement relationships of US tech companies. "Every company listed below has questionable revenue recognition methods. If all these transactions were presented graphically, it would be an incomprehensible and complex picture. In the future, people will see this as evidence of fraud, not a virtuous cycle," Burry stated. He added that the actual final demand is extremely small, and almost all customers are funded by their resellers.
This image features Nvidia and Intel. OpenAI, Oracle AMD, xAI, Microsoft Among the manufacturers listed, Nvidia is centrally located. The complex relationships between these companies include Oracle spending tens of billions of dollars to purchase Nvidia chips, a $300 billion cloud-related deal between OpenAI and Oracle , and Nvidia's planned investment of up to $100 billion in OpenAI. Michael Burry also questioned the timeframe for Nvidia's chips to generate profits, arguing that chips manufactured 3 to 4 years ago have already been sold, hence the longer depreciation period, but the fact that these products are in use does not guarantee profitability.
Orders are now shrouded in uncertainty.
Although Nvidia has recently refuted market claims about cyclical trading among tech companies and an AI bubble, its expressed confidence in the AI industry has not sufficiently resonated with investors, judging from its stock performance.
Nvidia CEO Jensen Huang refuted the AI bubble theory in a conference call last week, stating that Nvidia sees something different and is now in a virtuous cycle of AI. Nvidia CFO Colette Kress also refuted claims that Nvidia chips have a short lifespan, saying that chips from six years ago are still working at full capacity.
Reports indicate that Nvidia recently responded in a memo to Wall Street analysts, stating that its strategic investments represent a small percentage of its revenue, and that companies in its strategic portfolio primarily generate revenue from third-party customers rather than from Nvidia itself. The memo also emphasized the company's financial soundness. Nvidia stated that it differs significantly from companies involved in previous accounting fraud cases because its core business is economically sound, its reporting is comprehensive and transparent, and it highly values its reputation for integrity.
The changing AI landscape has cast a shadow of uncertainty over Nvidia's future order prospects, given its close relationship with OpenAI.
Nvidia has established a certain level of synergy with OpenAI and xAI through investments. Jensen Huang stated in a conference call last week that he believes the investment in OpenAI will translate into huge returns. However, Google released Gemini 3 last week, which achieved comprehensive leadership in almost all major benchmarks, posing a threat to OpenAI's dominant position. Reports indicate that OpenAI CEO Altman acknowledged in an internal memo that Google's recent progress in artificial intelligence may put temporary financial pressure on OpenAI.
If the competitive landscape for large-scale models changes, the computing power structure may also change. Google has made rapid progress in chip development; its fourth-generation TPU (Tension Processing Unit), released in April this year, boasts a peak single-chip computing power of 4614 TFLOPs, and Google's reliance on NVIDIA GPUs is relatively low. Although Google's TPUs are currently used for internal workloads and are not sold externally, if Google's own models are used more widely, there is a possibility that more AI workloads will run on Google chips in the future.
Nvidia also continues to face competition from other self-developed AI chip manufacturers . Competitive pressure from manufacturers. Compared to the versatility of Nvidia GPUs, ASIC (Application-Specific Integrated Circuit) chips, represented by Google's TPUs, have some advantages, such as theoretically higher energy efficiency and performance. OpenAI competitor Anthropic announced in October that it plans to use Google's computing power, deploying up to 1 million Google TPU chips to train its large AI model Claude. These TPU chips will be specifically used to accelerate machine learning workloads and are planned for deployment in 2026.
While the future of AI faces some skepticism, data still indicates strong current demand. Nvidia released better-than-expected Q3 FY2026 earnings last week, showing revenue up 62% year-over-year to $57 billion and net profit up 65% year-over-year to $31.9 billion. On the evening of November 25th, during the earnings call following the release, Alibaba... CEO Wu Yongming stated that there is unlikely to be an AI bubble for at least three years, and the pace of deployment of Alibaba Cloud AI servers and other products is seriously lagging behind customer demand.
To demonstrate to the market that AI demand will remain strong, Nvidia may need to provide more compelling evidence. Furthermore, to convince investors that the AI bubble won't burst, major tech companies and large-scale unicorns will need to work together. Then it needs to be proven that AI investment can be converted into sufficient revenue.
(Article source: CBN)
