DeepSeek, a Chinese AI startup founded by hedge fund chief Liang Wenfeng in May 2023, has been gaining attention for its innovative approach to AI technology. Unlike other models such as OpenAI’s ChatGPT and Alphabet’s Gemini, DeepSeek utilizes an open-source large language model that developers can customize to suit their specific requirements.
One of the key highlights of DeepSeek is its R1 model, which has been able to compete with OpenAI’s o1 model in areas like math, coding, and reasoning tasks, all while utilizing less advanced chips and at a significantly lower cost. For instance, it has been estimated that OpenAI’s o1 model costs $60 to produce a million tokens of output, while DeepSeek’s R1 can achieve the same output for just $2.19.
Despite impressing industry insiders with a detailed 22-page research paper outlining its model, DeepSeek has faced allegations from OpenAI of using a technique called distillation to develop its models, which involves training an AI model using larger, more capable ones. This accusation has raised concerns about potential plagiarism and credibility issues for the Chinese startup.
There is also uncertainty surrounding the actual cost of DeepSeek’s infrastructure, with reports suggesting that it only cost $5.6 million to train its V3 model using Nvidia H800s. However, analysts have questioned this figure, as it may not include essential expenses like research, architecture, or data. This makes it challenging to directly compare with U.S.-based AI models that have required significantly higher investments.
The implications of DeepSeek’s advancements on Nvidia and the wider AI industry remain unclear. While the company has shown promising improvements in efficiency, questions about the origin of its model and the true cost involved still linger. The launch of DeepSeek-R1 has been likened to a “Sputnik moment” by industry experts, hinting at potential geopolitical implications as well.
Investors should approach this development with caution, considering it as part of a broader technological evolution rather than a singular event. Even if DeepSeek’s open-source architecture leads to increased demand for AI chips, benefiting Nvidia in the long run, uncertainties surrounding the company’s practices and costs should be taken into account.
Why I’m not concerned about Nvidia’s future
Let’s not lose sight of the bigger picture when it comes to Nvidia. While the development of a competent AI chatbot like DeepSeek is noteworthy, our primary focus should be on achieving artificial general intelligence (AGI) for applications such as autonomous vehicles and robotics. It remains uncertain whether DeepSeek will significantly impact the trajectory towards AGI.
The technologies involved in AGI are immensely powerful and valuable, ensuring that the race towards this goal will persist. Major tech companies will continue investing billions in the necessary infrastructure to advance AGI. While efficiency is crucial, the true prize lies in technological leadership.
Although Nvidia experienced a setback with DeepSeek, the company has shown its ability to adapt in the past. Transitioning from a focus on video games to areas like cryptocurrency mining, AI, autonomous vehicles, 3D rendering, and more, Nvidia has proven its versatility. CEO Jensen Huang is widely recognized as a visionary in the industry, leading the company to innovate rapidly with its upcoming Rubin platform.
As the AI landscape evolves, Nvidia will adjust to changing market dynamics. Despite the impact of DeepSeek, the pursuit of AGI remains steadfast, as does Nvidia’s presence in this space.
Should you invest in Nvidia right now?
Before jumping into Nvidia stock, consider this: The Motley Fool Stock Advisor analyst team has identified what they believe to be the top 10 stocks for investors at present – and Nvidia didn’t make the list. The selected stocks are expected to deliver substantial returns in the years ahead.
Reflect on Nvidia’s absence from this list dated April 15, 2005. If you had invested $1,000 based on the recommendation at that time, you would now have $735,852*.
The Stock Advisor service offers investors a clear roadmap to success, featuring portfolio-building guidance, regular updates from analysts, and two new stock picks each month. Since 2002, the service has outperformed the S&P 500 by more than fourfold*.
For more information, visit the Stock Advisor website.
*Suzanne Frey, an executive at Alphabet, serves on The Motley Fool’s board of directors. Jeremy Bowman holds positions in Nvidia. The Motley Fool has positions in and recommends Alphabet, Microsoft, and Nvidia. Furthermore, The Motley Fool endorses the following options: long January 2026 $395 calls on Microsoft and short January 2026 $405 calls on Microsoft. The Motley Fool maintains a disclosure policy.
