China is afraid of ChatGPT because it gives uncensored answers
Chinese regulators have banned the country’s tech giants from providing access to the artificial intelligence chatbot ChatGPT over fears the tool would provide “uncensored answers” to politically sensitive questions.
This is reported by Nikkei Asia citing “people with direct knowledge of the matter”. Nikkei reports that Chinese regulators have ordered tech companies Tencent and Ant Group (a subsidiary of e-commerce giant Alibaba) not only to limit access to US-developed ChatGPT, but also to report to officials before launching their own competing chatbots.
The move is in line with the Chinese government’s tough approach to censorship and its rapid response to new technologies. Last month, for example, the country introduced new rules on the production of “synthetic content” – such as deepfakes.
Chinese tech giants have already been forced to censor other artificial intelligence apps, such as image generators. For example, one such tool launched by Baidu cannot generate image of Tiananmen Square.
Although ChatGPT is not officially available in China, it has caused a stir among Chinese internet users and the AI community, members of which are unhappy that China was not the first to develop the technology. Some cite the country’s strict technological regulation and censorship as an obstacle to creating such systems. The success of the United States in creating new chatbots is based in part on the large amount of training data pulled from the Internet, as well as the rapid launch and iteration of new models.
Nikkei reports that Chinese users have accessed ChatGPT through VPN services or third-party integrations into messaging apps such as WeChat, although WeChat developer Tencent has reportedly already banned some of these services.
Recently, China Daily, China’s largest English-language newspaper, warned on its social media that ChatGPT could be used to spread Western propaganda:
“ChatGPT has gone viral in China, but there is growing concern that the artificial intelligence could provide a helping hand to the US government in its spread of disinformation and its manipulation of global narratives for its own geopolitical interests,” said ChinaDaily reporter Meng Zhe.
In a YouTube video published by the outlet, another reporter asks ChatGPT about Xinjiang. The bot responds by citing “reports of human rights abuses against Uighur Muslims including mass internment in ‘re-education’ camps, forced labor, and other forms of persecution by the Chinese government” — a response that Xu-Pan describes as “perfectly in line with US talking points.”
Technology industry sources told Nikkei that the restrictions from Chinese regulators did not come as a surprise:
“Our understanding from the beginning is that ChatGPT can never enter China due to issues with censorship, and China will need its own versions of ChatGPT,” one tech executive told the publication.
Since ChatGPT launched last November, Chinese tech giants including Tencent, Baidu, and Alibaba have announced they are working on their own competing services. Search giant Baidu recently announced that its AI chat service ERNIE Bot will soon be integrated into its search services. However, it is unclear whether such a rapid development schedule will continue after Chinese regulators weigh the potential harm bots can cause.
Whatever happens next, it will be difficult for Chinese technology giants to navigate such restrictions. Limiting the training data for chatbots will reduce their capabilities compared to Western competitors, and even if the input is tightly controlled, users will still be able to receive unwanted responses for which companies will likely be held liable.