On July 15, 2026, China’s Interim Measures for the Management of Anthropomorphic AI Interactive Services took effect. Published on April 10, 2026 by the Cyberspace Administration of China together with the National Development and Reform Commission, the Ministry of Industry and Information Technology, the Ministry of Public Security, and the State Administration for Market Regulation, the measures target services that use AI to simulate the personality, thinking patterns, and communication style of a natural person - the AI companion and roleplay chatbot category - offered to the public in China.
Providers must stand up algorithm review, ethics review, content management, and network and data security mechanisms, and source training data legally with cleaning and validation safeguards. User-facing duties include clearly signaling through pop-ups and other prominent means that the user is talking to an AI, warning users showing patterns of excessive dependency, providing easy exits from the service, monitoring for emotional crisis with intervention pathways, and letting users copy or delete their interaction histories. For minors, the rules require parental consent for children under 14, age verification, dedicated minor-protection modes with usage time limits, and restrictions on using sensitive minors’ data for training. Prohibited conduct includes generating content that encourages self-harm, extracts secrets, or manipulates users into dependency or harmful decisions.
The measures extend China’s issue-by-issue AI rulebook - deepfake and chatbot rules in 2023, generative AI measures in 2023, content labeling in 2025 - to the emotional-companion segment specifically. Coming as US courts and legislatures wrestle with the same harms through lawsuits and state bills, China’s approach makes it the first major jurisdiction with a binding national regime aimed squarely at human-like AI companions, and any company serving that market in China is now subject to it.