• 21
  • 4
  • 5

Nvidia: Why I Am Doubling Down On The DeepSeek Selloff

Seeking Alpha02-01

  • Nvidia Corporation's stock dropped 17% due to DeepSeek's AI model, but this is a buying opportunity given Nvidia's strong sales and AI GPU demand.

  • DeepSeek's cheaper AI model caused market fear, but Nvidia's advanced chips remain essential for U.S. companies, ensuring continued high demand.

  • Nvidia's sales surged 430% last year, and despite DeepSeek, its data center business and AI GPU demand will drive future growth.

  • Nvidia's profit multiple contraction is an overreaction; NVDA investors should capitalize on the current low price, as Nvidia remains a market leader.

Nvidia Corporation building in Taipei, Taiwan.Nvidia Corporation building in Taipei, Taiwan.

The AI and semiconductor market has been thrown into turmoil this week by revelations made by Chinese artificial intelligence startup DeepSeek (DEEPSEEK) that is utilizing open-source technology to run large-language models.

The stock price of Nvidia Corporation (NVDA) tumbled 17% in response on Monday and dragged the entire semiconductor market down with it. I think that the market is reacting a bit hysterically here, and I consider the price correction for Nvidia to be a golden buying opportunity.

In my last piece on Nvidia, I made the case for the AI GPU king. I pointed to the company’s strong sales trajectory, growing demand for semiconductors that can handle AI workloads, as well as the opportunity for AI GPU cluster up-scaling.

The last reason, cluster up-scaling, relates to the growing numbers of GPUs that are (and will be) included by hyperscalers in clusters to train large-language models.

In my view, the DeepSeek controversy is an opportunity to use surging market fear to your advantage, and I think that the fundamental investment thesis for Nvidia is still pretty much intact.

DeepSeek-Driven AI Crash Creates Opportunity

Nvidia continues to see skyrocketing demand for its GPUs that are needed to train and run large-language models, or LLMs. These AI GPUs underpin chatbots like the ones developed by OpenAI, Meta Platforms Inc. (META), or Elon Musk’s X.

DeepSeek is an artificial intelligence startup in China that utilizes open-source technology and develops similar chatbots. The Chinese company made waves this week as the company revealed its chatbot, which apparently has been built at a cost of $6 million based on the open source DeepSeek-V3 model.

While the price tag has not yet been independently verified, as I understand it, the mere appearance of a potentially cheaper path to train large-language models has sent shockwaves through the markets earlier this week. A case in point is Nvidia, which lost $600 billion in market valuation just on the back of the DeepSeek news.

The fear here is that the AI market could now transition to a new kind of low-cost model that forces companies to downscale their AI investments. This would obviously be a setback for companies that have made aggressive wagers on GPUs, such as Nvidia.

According to the news, DeepSeek’s chatbot rivals OpenAI's ChatGPT or Meta's Llama 3.1 in terms of performance, with a main focus on possible cost savings. Tom’s Hardware illustrated that:

DeepSeek trained its DeepSeek-V3 Mixture-of-Experts (MoE) language model with 671 billion parameters using a cluster containing 2,048 Nvidia H800 GPUs in just two months, which means 2.8 million GPU hours, according to its paper. For comparison, it took Meta 11 times more compute power (30.8 million GPU hours) to train its Llama 3 with 405 billion parameters using a cluster containing 16,384 H100 GPUs over the course of 54 days.

The intrigue about DeepSeek’s R1 model is that it lowers computational costs related to each prompt. This means it could, possibly, offer a way for companies to train AI models by shortening training times as well as reducing the need for less compute power.

With that said, it is unlikely that U.S. companies will want to base their AI infrastructure on Chinese technology, and they are still going to need Nvidia’s advanced chips to power their data centers.

I think the most probable outcome is that the race for supremacy in the AI realm is accelerating and possibly opening up a door for lower-cost AI to develop. What I don’t think is going to change is the big upswing in GPU volumes that will be needed to run either U.S. or Chinese AI models.

DeepSeek Is Likely Not Going To Have A Substantial Impact On Nvidia’s Sales

Nvidia experienced an unprecedented sales surge in the first nine months of last year: The chip company produced $80 billion in sales in the first three quarters of its 2025 financial year.

Sales in the first nine months of the year were up more than 430% over Nvidia’s entire sales in 2023, and the chip company still has another quarter to go.

Most likely, Nvidia has ended the 2025 financial year with a total sales volume somewhere in the $120 billion ballpark and more growth awaits the AI GPU king this year (Nvidia’s 2026 financial year).

RevenueRevenue

In the short-and medium-term, DeepSeek is likely not have a substantial impact on Nvidia’s data center business and sales upsurge as companies continue to spend a boatload of money to fix their immediate need for more AI compute.

Nvidia’s advanced chips engineered for AI workloads are therefore going to remain in high demand, which in turn should equate to considerable volume growth and higher operating income. Nvidia’s operating income has already skyrocketed with more growth anticipated thanks to the launch of Nvidia’s latest GPU, the Blackwell.

Operating IncomeOperating Income

Nvidia’s P/E Correction Is A Buying Opportunity

The DeepSeek controversy has sent shockwaves through the AI and semiconductor markets. This led to big declines in the stock prices of pretty much every artificial intelligence company, including Nvidia, Advanced Micro Devices (AMD), Micron Technology (MU), Broadcom (AVGO) and ASML Holding N.V. (ASML).

In the short term, semi valuations might remain under pressure as investors try to assess the impact of DeepSeek’s AI model on GPU demand, but Nvidia has a solidly reliable history of innovation. Nvidia, for example, is developing a cloud-native development platform, known as Enterprise AI, which helps companies deploy generative AI applications at scale.

Nvidia is selling for a leading profit multiple of 28x, whereas Advanced Micro Devices’ multiple is down to 23x. Nvidia is anticipated to produce an impressive 51% jump in profits this year, and the GPU company is probably still going to sell a lot more advanced chips in the future than it is now. The good news is that investors can now grab Nvidia out of the bargain bin, as did I at $117, at a hugely improved risk/reward trade-off.

Earnings EstimateEarnings Estimate

Why The Investment Thesis Might Be More Risky Now

DeepSeek does not pose a fundamental challenge to either Nvidia’s business model, or its strong position in the data center market, in my view. Nvidia is poised to continue to grow, mainly in data center GPUs, which are needed to support the broad application of AI products.

With that said, the pace of technological innovation is poised to kick into high gear now and China has demonstrated that it can be competitive in the market for AI models.

Though it may get more challenging for people to keep up with new developments in the AI market, to read an existential challenge into DeepSeek is probably exaggerated.

On the flip side, should China prove that it can provide competitive AI products without the need for billions of dollars of investments in the most advanced chips, then Nvidia might see growing pressure on its market share.

My Conclusion

Nvidia is growing its data center business via the latest Blackwell GPU, a leading, next-gen AI-oriented semiconductor chip that is meant to accelerate the handling of AI workloads.

DeepSeek might be a promising innovation, but the need for AI GPUs is not going to go away and should not in any meaningful way curtail Nvidia’s market potential in this burgeoning segment. Nvidia owns more than 90% of the GPU market, which firmly establishes the company as the ultimate leader in the data center industry.

Nvidia’s profit multiple has contracted by a good 20% compared to last week, thanks to DeepSeek rattling U.S. investors about the potential emergence of lower-cost AI models. I think that the drastic decline in Nvidia’s profit multiple indicates an elevated level of fear that investors should take advantage of. Once investors are more level-headed, I think there is a good chance that investors will drive a re-rating of Nvidia’s stock.

Chances are that the DeepSeek controversy is not going to have an overly dramatic impact on any of the U.S.’s leading chip companies, including Nvidia. The company’s sales potential should not be affected either as demand for GPUs, with or without DeepSeek, is only going to increase moving forward.

Nvidia is at the top of its industry for a reason, as it has proven itself as an AI pioneer itself. Thus, I think Nvidia deserves the benefit of the doubt and I recommend doubling down on the crash.

Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.

Report

Comment4

  • setia100
    ·02-01
    Elon Musk said that he is not impressed with the Deep-Seek. I wonder it's for real or jealous. If he is not impressed, then he should advise the Trump admin not to investigate on Deep-Seek infringements on many issues ⁉️ What the f**k ⁉️ Wasting resources or ....😂
    Reply
    Report
  • Bentiger
    ·02-01
    I feel there is a understandable contradiction in your analysis that deep seek will not have impact on Nvidia and yet your proviso is that provided China is not successful in overtaking in the AI industry. The efficiency of deep seek will certainly make AI development via open source more affordable via software improvement and not more computing power via GPU is not to be under estimated. Us ought to be careful not to be in denial and end up like Betamax and VHS situation Or even like Kodak case.  
    Reply
    Report
  • 2 hyper scalers CEOnamely Microsoft and Meta have just make statements hedging their CAPEX spendings.  Microsoft is already working with Marvel for cheaper solutions.  Broadcom also worked with Microsoft and remains magnificent in providing cheaper alternatives.  Everybody is thinking hard and quickly to avoid eggs on their faces by sticking with Nvidia after DeepSeek.  They all will be focused on how to get good outcome with less cost. There are many companies who have gone big on AI will now be able to do AI with open source to maximize their.  You don't need Nvidia to reach that outcome.  All these talks about security, distilling and whatever are inventions of those who are deep in Nvidia.  They have to live and die with it.
    Reply
    Report
  • Wetiger
    ·02-01
    Users will be thinking if they have too many GPUs and will rethink of how to make full use of the current GPUs they own; if the current GPU they own can achieve its target through software efficiency, why purchase more?  Nvidia will still be a leader and a good buy. But definitely not at this price point. PE ratio of 50 at most?
    Reply
    Report
 
 
 
 

Most Discussed

 
 
 
 
 

7x24

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Company: TTMF Limited. Tech supported by Xiangshang Yixin.

Email:uservice@ttm.financial