DeepSeek, one of China’s leading artificial intelligence firms, made a rare public appearance at the World Internet Conference in Wuzhen, where its senior researcher Chen Deli expressed deep concerns about AI’s potential social fallout.
Chen said that while AI can greatly assist humans in the short term, it could replace most human work within the next two decades, creating a “massive challenge” for society. “Tech companies need to take on the role of defender,” he said, adding, “I’m extremely positive about the technology but view its impact on society negatively.”
The comments marked DeepSeek’s first major statement since January, when the company’s open-source model shocked the global AI community by outperforming several leading U.S. systems at a fraction of the cost.
DeepSeek has since become a flagship of China’s technological self-reliance, with state media framing its success as proof of national resilience amid U.S. chip export restrictions.
In recent months, DeepSeek has focused on refining its technology rather than public appearances. Its V3 model, launched in September, improved training efficiency and text processing. The company has also strengthened ties with Huawei and Cambricon, whose chips now support its models.
Analysts note that DeepSeek’s August announcement of a model optimized for Chinese-made chips briefly drove a rally in local semiconductor stocks, underscoring its growing strategic importance.
Chen’s remarks, however, highlighted a more introspective side of China’s AI boom—one that recognizes the risks of automation even as the nation races to challenge U.S. dominance in the field.
