China Cracks Down on AI Manipulated Videos Targeting Classic Films and Dramas
The Chinese government has intensified its efforts to control the use of artificial intelligence (AI) in media by targeting short video platforms.
Authorities are particularly focusing on videos that feature AI-generated twists on classic Chinese films and TV dramas, which they claim are distorting original stories and breaching intellectual property rights.
The National Radio and Television Administration (NRTA) recently called on popular short video platforms to identify and remove AI-created content that misrepresents traditional stories or cultural icons.
This move reflects Beijing's growing scrutiny of artificial intelligence use within the entertainment space.
The notice pointed to troubling examples such as scenes where the Monkey King, the legendary figure from Journey to the West, is shown riding a motorcycle—a bizarre twist to his story’s narrative.
Another instance includes depictions of Tang Sanzang, the monk central to Journey to the West, engaging in acts such as kissing a woman.
Such portrayals are not just untraditional; they clash with religious and cultural norms.
AI Videos Distort Classics
The NRTA’s crackdown extends to other examples of AI’s creative manipulation.
Reports highlight AI-altered versions of well-known dramas, such as The Legend of Zhen Huan transformed into a "gunfight action movie" and Dream of the Red Chamber reimagined as a martial arts thriller.
The changes are a stark departure from the original narratives and risk undermining the spirit of these iconic works.
The administration directly criticised such alterations, describing them as acts of cultural desecration.
According to the NRTA, these AI-generated videos are being used to chase online views and traffic, but they cross ethical boundaries by misrepresenting and distorting traditional narratives.
Furthermore, they may infringe upon intellectual property rights.
Specific measures have been outlined to address the problem.
Platforms are now required to conduct audits of AI-generated video content, with a deadline of 10 December 2024 to report their findings and efforts to remove misleading material.
Platforms must also adhere to strict review processes for AI-generated content, ensuring that any use of generative AI tools, such as AI editing technologies or effects, complies with government regulations.
These tools must include clear indications to viewers if content is AI-generated.
Legal Implications of AI Manipulation
Legal experts have raised concerns about the infringement risks of AI-generated adaptations.
Xia Hailong, a Shanghai-based lawyer, pointed out that such alterations could violate copyright by infringing the network distribution rights or adaptation rights of original creators.
Additionally, using the likeness of real actors without permission in AI-generated scenes could constitute violations of personal image rights and even reputational damage.
Concerns also focus on the widespread use of so-called “AI kiss” videos.
These films, where AI is used to artificially generate scenes of characters engaging in intimate interactions, have gone viral on short video platforms.
Investigations suggest that much of this content is created by unauthorised third-party accounts, often promoting sensationalised or lowbrow material that strays into inappropriate areas.
One such company developing these videos has already begun censoring certain keywords in response to public backlash and government pressure.
AI's Broad Influence on Popular Media
The popularity of platforms like Douyin (the Chinese equivalent of TikTok, owned by ByteDance) and Kuaishou has amplified the issue, with both companies developing their own AI technologies to create content.
ByteDance recently introduced a set of AI tools called Jimeng AI, Seaweed, and PixelDance, while Kuaishou has launched its own generative AI platform, Kling, earlier this year.
Both platforms have come under scrutiny as their tools contribute to the volume of AI-generated media flooding online networks.
China has adopted one of the strictest approaches globally to regulate AI, specifically addressing AI’s misuse in content creation.
Measures introduced in August 2023 under its "interim measures for the management of generative AI services" required companies to uphold values aligned with "socialist values."
Furthermore, in March 2024, Beijing launched campaigns to purge misleading AI-generated content and block accounts that exploit AI for rumour spreading or marketing manipulation.
The latest crackdown seeks to target what Beijing views as "unhealthy" uses of AI-generated media, aiming to prevent videos that distort history, culture, or traditional stories from spreading widely online.