One month after announcing its interim measures to regulate generative artificial intelligence, China has approved the first group of services propelled by large language models for public use.
China’s cyberspace authority has registered eight generative AI applications, including Baidu’s Ernie Bot, ByteDance’s Doubao, Sogou founder’s Baichuan, and chatbots based on the lab work of top research institutions such as Tsinghua University and Chinese Academy of Sciences.
After two decades of relatively laissez-faire tech policies, China is evidently more concerned with innovation control than development pace today. In recent years, China has enacted a number of licensing frameworks to reign in its formerly unrestricted tech industry, granting it greater control over the immense quantities of data that shape people’s ideas and lives.
To launch a game in China, for example, a company cannot simply publish it in app stores; rather, it must undergo an extensive filing procedure that includes stringent content evaluations.
Regulatory Hurdles and Market Potential: Generative AI’s Complex Landscape in China
The unpredictability of generative AI makes it an apparent target for the Chinese government, which is among the most proactive in regulating the rapidly evolving technology. The AI law stipulates that algorithms that can influence public opinion must be registered with the appropriate government body and that purveyors of generative AI services must obtain an administrative license to operate in the country.
The successful filings of the LLM chatbots could herald in a period of rapid growth and widespread adoption in China, several months after the tech titans unveiled their ChatGPT counterparts but restricted access to industry insiders. Baidu’s Ernie Bot ascended to the top of free iOS apps in China within a day of the filing announcement, giving the company’s share price in Hong Kong a modest boost on Thursday.