This article was originally published on Fool.com. All figures quoted in US dollars unless otherwise stated.
The Nasdaq Composite has led the market higher over the past couple of years, and while improving economic conditions have likely been the biggest catalyst, developments in the field of artificial intelligence (AI) run a close second. AI has been around for decades, but recent advancements have the potential to be a game-changer by streamlining mundane tasks and increasing productivity.
One of the biggest beneficiaries of this trend is Nvidia (NASDAQ: NVDA), which pioneered the graphics processing units (GPUs) that have become the gold standard for running AI systems. However, reports emerged recently that suggested a Chinese start-up had cracked the code for developing strong AI systems while using inferior chips and novel training processes. If true, this could spell bad news for Nvidia, whose stock has gained roughly 500% over the past two years (as of this writing), fueled by robust sales and even more impressive profits.
While the news is certainly concerning, I predict that once the panic subsides, Nvidia stock will soar to new heights. Here are three reasons why.
1. The devil's in the details
The biggest tech companies in the world have been buying Nvidia GPUs as fast as they can be made, and some reports suggest the supply is sold out for the next year. These AI-centric chips provide the computational horsepower needed to run these cutting-edge algorithms.
However, Chinese start-up DeepSeek claims it was able to achieve similar results using lower-performance processors and at a fraction of the cost. The company claims to have developed its latest system for just $5 million — when competing systems are estimated to cost between $100 million and $1 billion. These claims sent Nvidia — and many other AI-centric tech stocks — plunging. The popular narrative is that if AI models can be created using lower-quality and lower-cost chips, Nvidia's sales will fall off a cliff.
It's important to keep in mind that DeepSeek's claims may not stand up to scrutiny. In a note to clients this weekend, Bernstein analyst Stacy Rasgon wrote, "Did DeepSeek really 'build OpenAI for $5M?' Of course not." The analyst went on to note that many — if not most — of the inherent development costs and additional resources were not included in the price tag, calling the company's claims into question. To be fair, Rasgon suggests the cost to develop DeepSeek's R1 model was still probably less than existing systems, just not as cheap as they would have you believe.
2. An apples-to-oranges comparison
Much of the buildout of AI in data centers — which makes up the bulk of Nvidia's chip sales — has been necessary to meet the accelerating demand for AI. The vast majority of these use cases are business-related.
For example, Microsoft was quick to incorporate AI capabilities across a broad swath of its existing software-as-a-service (SaaS) offerings while also introducing Copilot, its suite of productivity enhancement tools. Palantir Technologies developed its Artificial Intelligence Platform (AIP), which has become the standard bearer for solving real-world business problems using a company's own data. A great many of the applications being developed are geared toward improving productivity for enterprise users.
Yet, as Wedbush analyst Dan Ives points out, "Launching a competitive large language model (LLM) for consumer use cases is one thing … launching broader AI infrastructure is a whole other ballgame, and nothing with DeepSeek makes us believe anything different." Put another way, while chatbots like ChatGPT and DeepSeek's R1 can be intriguing to consumers, they have more limited applications for enterprises, resulting in a much smaller opportunity. Ives goes on to say, "No U.S. Global 2000 is going to use Chinese start-up DeepSeek to launch their AI infrastructure and use cases."
So, while these developments in AI are certainly worth watching closely, they're unlikely to have any immediate near-term impact on the accelerating adoption of AI in the U.S. enterprise market.
3. Nvidia isn't a one-trick pony
It's easy to forget that it wasn't very long ago that the majority of Nvidia's revenue was from graphics cards used by gamers. In fact, as recently as early 2022, Nvidia generated more sales from gaming chips than from data centers. The advent of generative AI changed all that, but it helps illustrate an important point about Nvidia.
CEO Jensen Huang has proven extremely adept at positioning the company for the next big technological breakthrough, as evidenced by its success with AI. While the vast majority of Nvidia's current sales come from outfitting data centers for AI, Huang is already looking to the future.
The company has a deep bench of knowledge in the automotive realm, supplying hardware and software to carmakers working to develop self-driving cars. Nvidia also has a vast library of applications to help businesses build out copilots and agentic AI. It has also invested heavily to fuel advances in communications, robotics, and drones.
Nvidia has its fingers in many pies and we don't yet know what the company's "next big thing" will be. That said, and given Nvidia's history of innovation, it's likely just a matter of time.
Bonus round: All that opportunity at a discounted price
Finally, after Monday's shellacking, Nvidia stock is selling for around 27 times next year's expected earnings. This gives long-term investors the opportunity to buy shares in a company with a track record of innovation and a long runway for growth ahead — all at a discounted price.
This article was originally published on Fool.com. All figures quoted in US dollars unless otherwise stated.
