 "How DeepSeek's AI breakthrough could disrupt Nvidia & shake up tech stocks")
China’s DeepSeek, an emerging AI startup, has caught the attention of investors and analysts with its recent advancements, causing a noticeable drop in the stock market. On Monday, the Nasdaq fell 3.1 per cent, with Nvidia stock dropping 17 per cent — the largest decline of the day.
DeepSeek’s breakthrough centers on the launch of large language models (LLMs) that rival the performance of AI systems from industry leaders like OpenAI, Google, and Meta. Despite facing major obstacles, such as chip export restrictions, the company has delivered remarkable results, sparking speculation about the future direction of AI development. ALSO READ: DeepSeek's AI model impresses, but OpenAI's Sam Altman promises better
In December 2024, DeepSeek launched an LLM that demonstrated performance similar to OpenAI’s models. Then, in January 2025, the company unveiled another model, claiming it was trained at a fraction of the cost of its competitors. DeepSeek’s breakthrough, relying on open-source technology and innovative training methods, has left Wall Street intrigued by how the startup managed to deliver such results with limited resources.
DeepSeek’s impact on AI stocks
The success of DeepSeek has prompted analysts to question how the startup managed to develop competitive AI models without access to the same computing power that US tech giants use. DeepSeek’s ability to produce effective AI models with seemingly fewer resources is raising doubts about the future demand for high-end semiconductors, which have long been key to the success of companies like Nvidia.
In a recent report, Raymond James analyst Srini Pajjuri raised concerns that DeepSeek’s emergence could disrupt the demand for compute-intensive AI models. Nvidia, the dominant chip supplier for AI systems, saw its stock fall by 17 per cent, while other AI-related stocks, including Palantir, also experienced declines as investors re-evaluated the future of AI infrastructure.
“DeepSeek clearly doesn’t have access to as much compute as US hyperscalers and somehow managed to develop a model that appears highly competitive,” Pajjuri said, as quoted by Investor’s Business Daily.
Also Read
This has sparked discussions about the impact of DeepSeek’s success on the future of AI development and the demand for advanced semiconductors.
In a notable achievement, DeepSeek’s AI Assistant overtook OpenAI’s ChatGPT on the US App Store, highlighting the rising influence of the Chinese startup. Despite the uncertainty surrounding its access to restricted chips, DeepSeek’s rapid growth has caught the attention of US tech companies, which have invested heavily in AI infrastructure.
One of the strategies that sets DeepSeek apart from its competitors is its reliance on open-source technology. This approach contrasts with the proprietary models developed by companies like OpenAI, Google, and Meta. By offering its AI models to developers at a lower cost, DeepSeek is making AI more accessible and pushing for a more affordable alternative to the expensive models produced by the big players in the industry.
In contrast, companies like Meta and Microsoft have significantly increased their capital expenditures on AI, with Meta planning to invest up to $65 billion in AI infrastructure by 2025. DeepSeek’s low-cost, open-source approach could challenge the financial model of these established companies, which have long dominated the AI space.
DeepSeek’s AGI ambition
DeepSeek’s long-term vision is to develop Artificial General Intelligence (AGI)—AI that can perform tasks at or beyond human cognitive abilities. CEO Liang Wenfeng has made it clear that this is the company’s ultimate goal.
“We are studying new model structures to realise stronger model capability with limited resources,” Wenfeng said, adding, “Our destination is AGI.”
Impact on US tech
The rise of DeepSeek has also sent ripples through the US tech industry, raising concerns about potential profit pressures for major companies like Microsoft, Meta, Google, and Nvidia. US-based economist Ed Yardeni warned that these companies could face challenges if their AI services don’t generate returns that match their growing capital expenditures.
Yardeni also suggested that Nvidia could be impacted if customers begin to demand fewer and less advanced chips. However, he noted that US companies might be able to learn from DeepSeek’s efficient model and reduce their capital spending by designing AI systems that rely on cheaper chips.