A rush among big tech companies and startups to offer large language models (LLMs) has fueled the adoption of AI tools in China. As of June this year, the country has recorded more than 230 million users of AI products and services, according to a report by the China Internet Network Information Center (CNNIC), a state-run agency.
As of October, more than 600 million users had signed up to use one of the nearly 200 commercially available LLMs in China, according to a report. South China Morning Post.
According to the CNNIC report, 190 of the 309 AI companies that have registered LLM with the Cyberspace Administration of China have received approval so far. Companies need to get approval from the country’s internet regulator before making their LLM accessible to the public.
The AI product most used by local users is Chinese tech giant Baidu’s AI chatbot called Ernie Bot – which has a market share of 11.5 percent. Although OpenAI’s ChatGPT and Google’s Gemini models are not officially accessible in mainland China, they still command a market share of 7 percent and 3.8 percent, respectively.
According to a CNNIC report, nearly two-thirds of AI users in China use LLMs to answer questions while one-third of them rely on AI virtual assistants to generate slides and transcribe meetings.
The number of AI users in China suggests that adoption of LLMs in the country is growing and that domestic tech companies are looking to take on US-based AI giants such as OpenAI and Google.
Prominent Chinese startups such as Baichuan, Zhipu AI, Moonshot AI and MiniMax have launched their own AI models for use by consumers and enterprises. However, export controls and other restrictions imposed by the US on suppliers such as NVIDIA threaten to slow the progress of AI companies in China.