China has recently released new guidelines on generative AI services, which aim to limit their public use while encouraging industrial development. These guidelines, issued by the Cyberspace Administration of China (CAC), are set to take effect on August 15th. The regulations specifically target organizations that offer generative AI services to the public, while excluding entities that develop such technology for non-mass-market use.
Unlike the draft rules proposed in April, the new guidelines reflect a softened stance by the CAC. The regulations still emphasize that generative AI services must adhere to the core values of socialism, refraining from attempts to overthrow state power or the socialist system. However, the potential fines of up to 100,000 yuan ($13,999) for violations, as mentioned in the draft rules, have been removed in the updated guidelines.
It is important to note that China has been actively seeking to strengthen its generative AI capabilities, with the aim of surpassing the current dominance of the United States in this field. However, achieving this goal has not been without challenges for China. The country is known for its strict control over internet access and the dissemination of information within its borders. In fact, the government instructed its tech giants to refrain from accessing ChatGPT due to concerns about the chatbot providing “uncensored replies,” despite the tool not being available in China. Furthermore, the authorities cracked down on citizens using ChatGPT, even arresting an individual who allegedly used the chatbot to write fake articles.
The new generative AI rules in China also prioritize intellectual property rights related to training data. They explicitly prohibit the use of algorithms, data, platforms, or any other advantages to implement monopolies or engage in unfair competition. All training data utilized by service providers must come from sources approved and deemed legitimate by the government. Additionally, the guidelines require service providers to accept requests from individuals for the review or correction of information collected for AI models.
In addition to these restrictions, the Chinese government has expressed its intent to encourage the development of generative AI. This includes providing support for infrastructure and public training in this field. The government recognizes the potential of generative AI and aims to establish China as a leading provider in this sector.
These guidelines reflect China’s ongoing efforts to balance technological advancements with the control and regulation of AI services. The country is keen on harnessing the benefits of generative AI while ensuring it aligns with its societal values and maintains control over information dissemination. By imposing limitations on public use while promoting industrial development, China aims to strike a delicate balance between innovation and regulation in the field of generative AI.
