China proposes strict AI rules to protect children
Rising AI usage worldwide, including platforms with millions of users, has prompted China to tighten oversight and ensure chatbots operate safely and responsibly.
China has proposed stringent new rules for AI aimed at protecting children and preventing chatbots from providing advice that could lead to self-harm, violence, or gambling.
The draft regulations, published by the Cyberspace Administration of China (CAC), require developers to include personalised settings, time limits, and parental consent for services offering emotional companionship.
High-risk chats involving self-harm or suicide must be passed to a human operator, with guardians or emergency contacts alerted. AI providers must not produce content that threatens national security, harms national honour, or undermines national unity.
The rules come as AI usage surges, with platforms such as DeepSeek, Z.ai, and Minimax attracting millions of users in China and abroad. The CAC supports safe AI use, including tools for local culture and elderly companionship.
The move reflects growing global concerns over AI’s impact on human behaviour. Notably, OpenAI has faced legal challenges over alleged chatbot-related harm, prompting the company to create roles focused on tracking AI risks to mental health and cybersecurity.
China’s draft rules signal a firm approach to regulating AI technology as its influence expands rapidly.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!