A Chinese artificial intelligence startup, DeepSeek, has stunned the global technology industry and disrupted financial markets with the release of its AI model, R1. The revelation has led to a historic selloff in US tech stocks on Monday, raising questions about America’s long-held dominance in AI innovation and investment.
A “Sputnik Moment” for AI
DeepSeek’s R1, a ChatGPT-like model, offers comparable capabilities to its Western counterparts but at a fraction of the development cost. The company claims to have spent just €5.1 million to build its model, a stark contrast to the billions invested by US firms like OpenAI, Google, and Meta.
The announcement has been described as AI’s “Sputnik moment” by venture capitalist Marc Andreessen, evoking the 1957 satellite launch that spurred the space race. The R1 model quickly became the most downloaded free app on Apple’s US App Store over the weekend, highlighting its rapid acceptance and appeal.
The news prompted a dramatic reaction on Wall Street. Nvidia, the leading supplier of AI chips, saw its stock plunge by 17%, erasing €536.8 billion in market value—the largest single-day loss in US stock market history. Other major tech companies, including Meta and Alphabet, experienced significant declines, while Nvidia competitors and energy firms tied to AI infrastructure also faced sharp losses.
A Challenge to US AI Dominance
DeepSeek’s emergence has shaken the confidence that US companies, backed by abundant data centres and advanced semiconductor technology, hold a commanding lead in AI development. Just days before, OpenAI CEO Sam Altman, Oracle founder Larry Ellison, and President Donald Trump had unveiled “Stargate,” a €455 billion joint venture aimed at bolstering AI infrastructure in the US.
However, DeepSeek’s breakthrough has disrupted long-standing assumptions about the capital and infrastructure required to develop cutting-edge AI. The startup’s use of existing technology and open-source software has demonstrated that innovation can thrive outside the high-cost models traditionally employed by US firms.
Analysts are now scrutinising the company’s financial claims. Gene Munster, a veteran industry observer, noted that DeepSeek’s low development costs are “hard to believe,” raising questions about whether the company is receiving subsidies or leveraging undisclosed resources. Reports suggest that DeepSeek’s founder, Liang Wenfung, had stockpiled GPUs (graphics processing units) for years, enabling the company to sidestep US restrictions on high-powered AI chips.
Market Implications
The impact of DeepSeek’s entry extends beyond the technology sector. Energy stocks tied to the AI boom, including Constellation Energy and Vistra, fell sharply as the market reconsidered the necessity of energy-intensive AI infrastructure projects. Futures for natural gas dropped 5.9%, and crude oil prices fell more than 2%, reflecting a broader reassessment of energy demand tied to AI.
The selloff highlights a shift in investor sentiment. Keith Lerner, an analyst at Truist, remarked, “The DeepSeek model rollout is leading investors to question the lead that US companies have and how much is being spent.”
A Turning Point for the AI Race
DeepSeek’s success is being hailed as a significant achievement for China, showcasing its ability to compete in the high-stakes AI race despite US efforts to restrict access to advanced technologies. The company’s R1 model has sparked renewed interest in Chinese AI firms, which have been undervalued due to geopolitical tensions and global economic uncertainties.
Charu Chanana, chief investment strategist at Saxo, noted, “DeepSeek’s rise could spark renewed investor interest in undervalued Chinese AI companies, providing an alternative growth story.”
While DeepSeek’s accomplishment is significant, its long-term impact remains uncertain. The R1 model is designed for consumer use and has yet to prove its viability for more complex industrial applications. US companies, meanwhile, are expected to respond with advancements that could maintain their competitive edge.
Sam Altman acknowledged DeepSeek’s achievement in a late-day post on Monday, calling it “impressive” but expressed confidence in OpenAI’s ability to develop superior models. “We will obviously deliver much better models,” Altman wrote, adding that the competition is “invigorating.”
A Wake-Up Call
The emergence of DeepSeek has been a “wake-up call” for the US technology sector, according to President Trump. He emphasised the importance of innovation in reducing the costs of AI development, calling DeepSeek’s success a “good thing” for the industry.
DeepSeek’s disruptive entry may also prompt a reassessment of US export controls on AI technology. The company reportedly used Nvidia’s H800 chips, which comply with US restrictions, raising questions about the effectiveness of current policies in curbing China’s technological advancements.
