Javier Lacort
Senior WriterI write long-form content at Xataka about the intersection between technology, business and society. I also host the daily Spanish podcast Loop infinito (Infinite Loop), where we analyze Apple news and put it into perspective. LinkedIn
In just two weeks, an unknown 40-year-old Chinese engineer has shaken up Silicon Valley and dominated the global conversation about AI, even triggering a global stock market earthquake.
Liang Wenfeng has accomplished something that once seemed impossible. He’s developed an AI model that rivals OpenAI at a fraction of the cost. In other words, he’s launched a challenge to U.S. dominance in the AI field.
Liang isn’t your typical tech entrepreneur. Born in Zhanjiang in Southern China, he studied electrical engineering at Zhejiang University and earned brilliant grades.
In 2015, he co-founded High-Flyer, a quantitative investment fund that went on to manage more than $13 billion using machine learning algorithms to trade in the stock market.
What distinguishes Liang from others is his unconventional career path. While most Chinese AI companies have focused on commercializing products, he’s prioritized pure research.
“In the past thirty years, [China’s tech industry] has only emphasized making money, and ignored innovation. Innovation is not solely driven by business, it also needs curiosity and a desire to create,” Liang told the Chinese publication Waves.
In 2021, just before the U.S. imposed restrictions on China, his vision took shape when he began stockpiling thousands of Nvidia chips for an unnamed project. Two years later, he founded DeepSeek with just over $1 million in seed capital. Currently, local media report that DeepSeek only has 140 employees, which is about 10% of the size of OpenAI.
DeepSeek’s success has elevated Liang’s status in China. On Jan. 20, he attended a private symposium with Prime Minister Li Qiang, where he was the youngest participant.
Liang has quickly emerged from relative obscurity in the AI field to become a central figure in global tech discussions. This has prompted some to question whether DeepSeek could have developed its flagship models, V3 and R1, using only the resources the company has publicly acknowledged.
Scale AI CEO Alexandr Wang expressed skepticism in comments to CNBC, suggesting that DeepSeek likely had access to a greater supply of chips than it could openly admit due to commercial restrictions. Anthropic CEO Dario Amodei offered a more sympathetic perspective, showing respect without being condescending.
For Liang, the aim extends beyond merely competing with Silicon Valley. In an interview with Chinese media outlet 36Kr, he explained his vision for China to "gradually transition" from being a beneficiary to a contributor in the AI sector.
“What we see is that Chinese AI can’t be in the position of following position,” he said. “We often say that there is a gap of one or two years between Chinese AI and the United States, but the real gap is the difference between originality and imitation.”
Liang’s low-key demeanor and scruffy appearance contrast sharply with those of OpenAI CEO Sam Altman and Meta CEO Mark Zuckerberg. Some of his peers describe him as a pragmatic leader driven more by curiosity than by the pursuit of wealth or fame, which aligns with what we’ve seen so far.
The DeepSeek CEO’s emphasis on research over commercial applications reveals an underlying aspect of his personality: He prioritizes a quest for long-term knowledge over immediate profits.
This focus may well be what has changed China’s position in the global AI race.
Image | Tim Reckmann | X | Xataka | DeepSeek
Log in to leave a comment