The Cyberspace Administration of China (CAC) has released a draft of the “Interim Measures for the Administration of Anthropomorphic Interactive Services Using Artificial Intelligence” for public consultation, aiming to regulate AI services that simulate human personality traits and provide emotional interaction, Xinhua News Agency reports. The public can submit feedback via email or mail by January 25, 2026.

The draft, issued under China’s Civil Code, Cybersecurity Law, Data Security Law, and other relevant regulations, seeks to ensure the healthy development and standardised use of AI-powered anthropomorphic services while safeguarding national security and public interests. It applies to AI products and services that engage users through text, images, audio, and video.

Key provisions require providers to clearly inform users that they are interacting with AI and to issue dynamic pop-up reminders if users show signs of overdependence or prolonged use. Providers must implement comprehensive security and data management measures, covering algorithm review, ethics evaluation, personal information protection, and emergency response.

Special attention is given to vulnerable groups. Minors must have guardian consent, minor-specific safety modes, usage limits, and real-time alerts for guardians. Elderly users must be guided to set emergency contacts, and providers are prohibited from simulating familial relationships.

The draft prohibits AI services from generating content that threatens national security, undermines social order, spreads rumours, promotes violence, or harms users’ physical or mental health. Providers must also conduct regular risk assessments, maintain complaint mechanisms, and cooperate with supervision by CAC and other authorities.