Select Language:
Chinese social media platform Xiaohongshu announced today that it will begin restricting the spread of artificial intelligence-generated content that isn’t properly labeled. This decision comes after highly realistic videos created by ByteDance’s Seedance 2.0 model stirred widespread controversy.
The initiative responds to numerous user complaints about AI-generated videos impersonating public figures and producing deceptive content, which has raised concerns over portrait rights and online trust. It also coincides with increased regulatory scrutiny of unmarked and misleading AI content by Chinese authorities.
An official statement from Xiaohongshu explained that some AI-created posts had failed to include proper labels, infringing on the rights of public figures and eroding user confidence. The platform emphasized its commitment to enhancing its ability to detect AI-generated material and encouraged users to proactively mark such posts. Content lacking clear AI disclosure will face distribution restrictions.
Recently, many videos generated by ByteDance’s Seedance 2.0 have circulated widely online. These clips mimic real footage very closely and replicate the appearances of certain public figures or TV and film characters, garnering substantial attention and sparking debates about potential portrait rights violations.
On February 10, the Seedance 2.0 team issued a statement saying they are making urgent improvements based on public feedback and have temporarily halted the feature that allows users to create videos using real images or videos as references.
China’s cyberspace regulator has also stepped up efforts against AI misuse. Yesterday, they announced that over 13,420 accounts using unmarked AI-generated content to deceive or mislead users had been investigated, leading to the removal of more than 543,000 pieces of violating content.
The regulator also plans to launch a month-long campaign focused on combating the inappropriate use of AI to produce vulgar or violent material, spread false information, or generate malicious hype.
ChatGPT
Perplexity
Gemini AI
Grok AI
