15

DeepSeek, a Chinese artificial intelligence app, has outpaced ChatGPT to become the top free app on Apple’s App Store in the US, UK, and China. Since its launch in January, DeepSeek has gained massive popularity, challenging the belief that the United States dominates the AI industry.

The app runs on the open-source DeepSeek-V3 model, developed for less than $6 million. This low-cost approach stands in sharp contrast to the billions spent by its competitors. However, some in the AI community have questioned this cost claim. Earlier this month, the company introduced DeepSeek-R1, which it claims matches OpenAI’s latest models in tasks like mathematics, coding, and natural language reasoning.

Marc Andreessen, a Silicon Valley venture capitalist and advisor to Donald Trump, called DeepSeek-R1 “AI’s Sputnik moment,” referencing the Soviet Union’s groundbreaking 1957 satellite launch.

Chinese Innovation Amid US Restrictions

Advanced chips are vital for training AI models like DeepSeek and ChatGPT. However, US restrictions since 2021 have limited China’s access to advanced chips. In response, Chinese developers have shared resources and experimented with new techniques, creating AI models that require less computing power and cost significantly less.

This innovation has disrupted the global AI landscape. US-based AI giants like Nvidia, Microsoft, and Meta saw their stock prices drop after DeepSeek’s success. European companies also felt the impact, with ASML’s share price falling over 10% and Siemens Energy’s plunging by 21%.

Fiona Cincotta, senior market analyst at City Index, noted that a low-cost Chinese AI model surprised the market and raised concerns about profitability for rivals heavily invested in expensive infrastructure. Vey-Sern Ling, a Singapore-based technology equity advisor, warned the BBC that DeepSeek might derail the investment case for the entire AI supply chain.

Despite DeepSeek’s success, Citi analysts caution that US companies may still hold an edge due to access to advanced chips. Meanwhile, a consortium of US tech firms and investors recently announced The Stargate Project, committing $500 billion to AI infrastructure in Texas.

DeepSeek’s Origins and Leadership

DeepSeek was founded in 2023 by Liang Wenfeng in Hangzhou, China. A graduate in information and electronic engineering, Liang also founded the hedge fund that financed DeepSeek. He reportedly amassed a large supply of Nvidia A100 chips—banned from export to China—which he paired with cheaper, lower-end chips still available for import. Experts estimate Liang’s chip collection includes around 50,000 units.

Liang has gained attention on the global stage. He was seen meeting with Chinese Premier Li Qiang and, in a 2024 interview with The China Academy, expressed surprise at the focus on pricing.

“We didn’t expect pricing to become such a sensitive issue,” Liang said. “We were simply calculating costs and setting prices accordingly.”

DeepSeek’s rise shows how innovation and adaptability can reshape the AI industry, challenging long-standing assumptions about dominance and cost.