Chinese social media platforms remove AI-generated harmful content targeting minors
Chinese social media platforms Douyin and WeChat removed hundreds of thousands of AI-generated videos that harmed minors, including altered animations containing violent and sexual content, and manipulated classic literature and historical narratives. The platforms identified cases of remote sexual harassment and content promoting harmful behaviors to children, taking enforcement action against thousands of accounts.
What happened
Since 2026, Douyin has removed approximately 400,000 pieces of content violating rules on minors, including AI-generated content that maliciously altered children's animation characters such as 'Ultraman' and 'Bear Infested' by inserting bloody, violent, and crude elements to create 'sinister children's content.'
WeChat reported removing over 8,000 AI-altered videos since 2026, including 4,376 in January and 3,956 in February, with creators using AI to rewrite classic literature such as 'Romance of the Three Kingdoms' and 'Water Margin,' distorting original content and demeaning historical figures.
Reported impact
Douyin identified multiple cases of remote sexual harassment of minors, including users impersonating 18-year-old males to lure minors into sending private videos, with the platform assisting police in arresting 8 criminal suspects.
WeChat noted that AI-altered content distorted historical figures and hero narratives, corrupted animation characters familiar to minors, misled youth values, hurt national sentiment, and eroded mainstream cultural consensus.
Index analysis
The platforms attribute the proliferation of harmful content to users employing increasingly covert tactics, including encrypted terminology and coded language to promote extreme fasting and drug abuse, content easily imitated by minors.
Governance controls
Douyin has taken enforcement action against 1,030 violating accounts through silencing and bans, processed 529 accounts engaged in producing and spreading AI-altered animation content, and removed related videos; the platform reports it will upgrade technical identification and review capabilities to improve detection of hidden and evolving violations.
WeChat has banned over 1,200 accounts for AI-based impersonation of public figures and is implementing long-term governance measures for AI-altered video content with increased enforcement against violating accounts and content.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.