(Corrects in paragraph 16 to province of Guangdong, not metropolis)
By Eduardo Baptista
BEIJING (Reuters) -Liang Wenfeng, the 39-year-old founding father of Chinese language AI startup DeepSeek, has within the matter of weeks change into the face of China’s tech trade and its hope of overcoming an ever-tightening noose of export controls imposed by the US.
Liang had saved a particularly low profile till Jan. 20, when he was one among 9 people requested to present a speech at a closed-door symposium hosted by China’s Premier Li Qiang.
He gave two uncommon media interviews to Chinese language media outlet Waves final 12 months and in 2023, however aside from that has stayed largely out of the general public eye. DeepSeek didn’t reply to a request for an interview.
On the symposium, the millennial’s youthful look contrasted with the grey-haired teachers, officers and state-owned conglomerate heads sat round him, photos and video printed by Chinese language broadcaster CCTV confirmed.
However the reality Liang was invited to share his opinions on Chinese language authorities coverage highlights Beijing’s recognition of DeepSeek’s position in doubtlessly upending the worldwide AI order, in China’s favour.
DeepSeek launched a free AI assistant final week that the agency says makes use of much less information at a fraction of the price of present companies, triggering a worldwide selloff in tech shares.
Final 12 months, Baidu CEO Robin Li spoke at an identical symposium chaired by the Chinese language premier. Li, who introduced China’s first ChatGPT rival in March 2023, mentioned in an interview that very same 12 months China would by no means recreate Microsoft-backed OpenAI’s success and that Chinese language corporations ought to concentrate on making use of present AI fashions for business functions.
Underneath Liang’s management, DeepSeek intentionally prevented app-building. As an alternative, it concentrated analysis expertise and sources on making a mannequin that would match, or higher OpenAI, and it hopes sooner or later to proceed specializing in cutting-edge fashions that will probably be utilized by different firms to construct client and enterprise-facing AI merchandise.
Liang’s strategy stood out in a Chinese language tech trade that was used to taking improvements from overseas, from smartphone apps to electrical automobiles, and rapidly scaling them up, typically a lot quicker than the nations the place the innovations have been first made.
“China’s AI cannot be within the place of following without end. We frequently say that there’s a hole of 1 or two years between China’s AI and the US, however the true hole is the distinction between originality and imitation,” Liang mentioned in an interview with Waves in July final 12 months.
