China's cyber regulator has released draft rules to govern AI services that simulate human personalities and engage users emotionally. The proposed regulations aim to strengthen safety and ethical standards for AI products interacting through text, images, audio, or video. Key provisions include requiring providers to warn users about excessive use, intervene in cases of addiction or extreme emotions, and ensure data security and personal information protection. The rules also prohibit AI from generating content that endangers national security or promotes violence and obscenity.