China has narrowed the artificial intelligence (AI) development gap with the United States to just three months in certain areas, according to Lee Kai-fu, the CEO of Chinese AI startup 01.AI.
Lee, a key figure in the global AI landscape and former head of Google China, highlighted that firms like DeepSeek have found innovative ways to use chips and apply algorithms more efficiently, propelling China closer to catching up with the U.S.
Lee made these remarks in an interview with Reuters, noting that the launch of DeepSeek’s AI reasoning model in January marked a significant milestone. The model, trained with less advanced chips, was cheaper to develop compared to Western counterparts, challenging the assumption that U.S. sanctions were stalling China’s AI sector. "Previously, the gap was six to nine months in many areas. Now it’s probably just three months behind in core technologies, and actually ahead in some specific areas," Lee said.
He added that Washington’s semiconductor sanctions had presented short-term challenges for Chinese firms but had also spurred innovation. Lee pointed out how Chinese companies, like DeepSeek, had developed new algorithms under constraints, including a novel approach to reinforcement learning. This method shows users the reasoning behind AI decisions, a capability first developed by OpenAI but not yet available to users in its current form.
Since the launch of OpenAI’s ChatGPT in late 2022, China’s tech sector has rapidly entered the generative AI race, but DeepSeek’s breakthrough has marked a key turning point. Until DeepSeek’s release, many Chinese tech leaders had expressed concerns about their lag behind Western rivals.
Lee, who founded 01.AI in March 2023, emphasized that investing in proprietary models was a "courageous" move for startups in a market dominated by well-established giants and fast-evolving open-source alternatives. While many of China’s top tech companies, including Baidu, Alibaba, and ByteDance, are building foundational models, 01.AI is focusing on practical AI applications, such as software solutions that help clients deploy AI technology effectively.
Earlier this month, 01.AI launched Wanzhi, a software platform designed to assist enterprises in adopting AI solutions. The company has already begun generating revenue, and Lee expects substantial growth in 2025, projecting that revenue will significantly exceed the $15 million achieved in 2024.
Read next
18:30
DeepSeek
U.S. Commerce Department bureaus have reportedly issued a blanket ban on the Chinese AI model DeepSeek for use on government-furnished equipment, according to sources and a mass email seen by Reuters.
17:00
DeepSeek
Chinese retail investors are increasingly turning to DeepSeek and other artificial intelligence tools to navigate the stock market, marking a significant shift from last year’s government crackdown on computer-driven quantitative trading.
21:00
DeepSeek
The Trump administration is reportedly considering a ban on the Chinese AI chatbot DeepSeek from U.S. government devices over national-security concerns, according to a source familiar with the matter.
20:30
xAI
Elon Musk-owned artificial intelligence firm xai has acquired a 1-million-square-foot property in southwest Memphis, the Greater Memphis Chamber reported on Friday.
22:00
DeepSeek
Chinese AI startup DeepSeek has stirred conversation in the tech community by claiming that its AI models could achieve a “cost profit margin” of 545% under ideal conditions.
What is your opinion on this topic?
Leave the first comment