China’s Cyberspace Administration has drafted provisional regulations to govern anthropomorphic AI interaction services, prohibiting the generation and dissemination of content that endangers national security, damages national honor and interests, or spreads rumors that disrupt economic and social order.
According to the official website of China’s Cyberspace Administration, the draft “Interim Measures for the Management of Anthropomorphic Artificial Intelligence Interaction Services” was released for public consultation on December 27th. The proposed regulations aim to promote the healthy development and standardized application of anthropomorphic AI interaction services while safeguarding national security, social public interests, and the legitimate rights of citizens, legal entities, and other organizations. The measures advocate for innovative development of these services while implementing prudent and classified supervision to prevent abuse and loss of control.
The draft outlines eight prohibited activities for providers and users of anthropomorphic interaction services. These include generating or spreading content that harms national security, damages national honor and interests, undermines ethnic unity, conducts illegal religious activities, or spreads rumors disrupting economic and social order. Additionally, the regulations prohibit content promoting obscenity, gambling, violence, or criminal instruction, as well as content that insults or defames others and infringes upon their legitimate rights.
The measures also forbid services that provide false promises significantly affecting user behavior or damage social relationships. Further restrictions address activities that harm users’ physical health through encouraging or glorifying suicide and self-harm, or damage psychological health through verbal violence and emotional manipulation. The regulations prohibit inducing unreasonable user decisions through algorithmic manipulation, information misleading, or emotional traps, as well as soliciting classified sensitive information and other violations of laws and regulations.
Source: Central News Agency (Taiwan), December 28, 2025
https://www.cna.com.tw/news/acn/202512280074.aspx