Where will Nvidia’s stock be in 3 years?

If you have been held to Nvidia (NVDA -3.67%)) Stock for three years, you probably laugh to the bank. A bet of $ 10,000 made at the start of 2022 is worth nearly $ 55,000 today – an impressive efficiency of 450%. That said, past performance does not guarantee future results. And the dynamics that helped Nvidia achieve such remarkable performance could change considerably.

The new large -language Chinese model (LLM) Deepseek is the elephant of the room. He would have shown that advanced AI models can be constructed without scandalous costs, in contrast that hit with the strategies used by its American rivals, like Openai, AlphabetAnd Meta-platforms. Let’s explore how this story could develop over the next three years and beyond.

This is not only on Deepseek

Although it is easy to blame Deepseek for all the challenges of the AI ​​industry, it could miss the overview. The basic problem is that LLMS training is too expensive compared to their profit potential. While the emergence of a radically cheaper Chinese rival has brought this problem in the dominant current, industry observers have been discussing it for months.

In June 2024, Goldman Sachs has published a report entitled “Gen Ai: Too much expenses, too few advantages”. The document suggested that the $ 1 billions in capital expenses The giants that technologies should pour out in AI equipment and generative development are shortly due to the inherent limitations of technology. He argued that LLM may not be ideal for solving complex problems that can justify their development costs.

For Nvidia, this problem is distant because it focuses on the Pick-And-Pivot side Graphic processing units (GPUS) that Cloud Computing companies use to build their IA calculation power (which is then rented to consumer-oriented start-ups). However, as these start-ups continue to fail or burn money (Openai lost $ 5 billion in 2024), this can be a matter of time before the chip demand begins to slow down.

Deepseek makes the problem too obvious to ignore

While the AI ​​bulls were content to ignore the warning signs of unsustainable industrial expenditure, Deepseek made the problem too obvious to ignore. With its launch in January, the company’s open source generative Chatbot quickly became the most downloaded free application in the United States AppleApp of the app – beating its American rival Chatgpt.

More importantly, the DEEPSEEK R1 model has similar performance on O1 Chatgpts despite its development for $ 6 million using Nvidia H800 chips, according to its developers. OPENAI says that Deepseek may have used its owner models to form R1 in a process called “Distillation”, and the White House AI and the Tsar Crypto David Sacks believe “that it is possible” that IP S flight ‘is produced during its development.

In the coming years, The US government may be able to Intervene on behalf of American American companies by more restricting flea exports to China (a decision that could affect Nvidia sales in this country) or try to undermine Deepseek according to the application of intellectual property or National security problems, as has been done in Tiktok.

But ultimately, it seems that the cat has come out of the bag, and even if the United States manages to eliminate Deepseek, other cheaper LLM could get up to take its place.

The already questionable economy behind the expenses of the Big Tech IA fleas has become equal Stronger To justify.

Where will Nvidia be in the next 3 years?

In the coming years, investors should expect Nvidia growth and margins to decrease as industry participants realize that they do not need their latest equipment (and the more expensive) to develop LLM and other AI generating applications. But although the exponential growth trajectory of the action will probably end, investors should not expect an accident.

With a Price in the future Several NVIDIA shares are surprisingly affordable for a company that increased 94% sales during its last quarter. The low evaluation suggests that some of Nvidia’s long -term challenges are already assessed.

Randi Zuckerberg, former Director of Development of the Facebook and Sister of the CEO of Meta Platforms, Mark Zuckerberg, is a member of the board of directors of Motley Fool’s. Suzanne Frey, director of Alphabet, is a member of the board of directors of Motley Fool’s. Will Ebiefung has no position in the actions mentioned. The Motley Fool has positions and recommends Alphabet, Apple, Goldman Sachs Group, Meta Platforms and Nvidia. The Motley Fool has a policy of disclosure.