Key takeaways
- China’s National Radio and Television Administration (NRTA) launched a special two-month governance campaign against AI-generated micro-dramas, running from November 1 to December 31, 2026, as reported by IT之家 on November 6, 2026.
- The campaign targets two categories of problems: content creation (e.g., soft pornography, extreme polarization, ‘盗脸’ (unauthorized face cloning) and ‘偷声’ (unauthorized voice replication)) and algorithmic promotion (e.g., repetitive rebranding of same content, low-diversity homepage recommendations).
- It builds on prior NRTA actions against low-quality and infringing micro-dramas — but marks the first publicly confirmed regulatory intervention explicitly focused on AI’s role in amplifying harmful or homogenized content.
- The campaign mandates compliance with existing备案 (filing and review) requirements for special题材 content (e.g., historical, political, religious, military, medical, legal themes), and introduces explicit norms around synthetic media consent — notably ‘盗脸’ (unauthorized face cloning) and ‘偷声’ (unauthorized voice replication).
- As of November 6, 2026, no enforcement metrics, penalties, or platform-level compliance data were disclosed in the source or in contemporaneous official releases.
What happened
On November 1, 2026, China’s National Radio and Television Administration (NRTA) initiated a special two-month governance campaign targeting AI-generated micro-dramas. As confirmed by IT之家 on November 6, the campaign responds to observed deterioration in content quality and ethics following rapid adoption of generative AI tools by micro-drama production studios and platforms. Unlike earlier NRTA efforts targeting copyright infringement or overtly illegal content, this round explicitly identifies AI not as a neutral tool, but as an enabling vector: one that lowers production barriers so drastically that it accelerates both homogenization and the resurgence of banned tropes — including soft-pornographic framing, family-ethics violations, and algorithmically optimized ‘vulgar clickbait’. The campaign is structured across two axes: content creation and platform distribution. On the creation side, it reinforces mandatory pre-release filing for special题材 content and introduces enforceable boundaries around synthetic identity — banning unauthorized use of real people’s faces and voices. On the distribution side, it challenges how platforms curate and amplify content: prohibiting deliberate rebranding (e.g., renaming identical scripts to evade detection) and demanding algorithmic diversity in homepage and ranking placements.
Why it matters
This intervention signals a decisive shift in China’s AI governance logic — from regulating what is produced to regulating how production scales. The NRTA’s move confirms that regulators treat algorithmic amplification as inseparable from content generation: a low-quality script becomes socially consequential only when pushed to millions via recommendation engines trained on engagement, not merit. By mandating diversity in homepage curation and penalizing ‘repackaged repetition’, the campaign effectively treats recommender systems as co-authors — subject to editorial accountability. That represents a novel extension of China’s existing ‘platform responsibility’ doctrine, previously applied to news feeds and e-commerce listings, now extended into narrative entertainment. It also reflects growing recognition that AI’s greatest near-term societal risk in China may not lie in job displacement or military use — but in the erosion of shared cultural thresholds, accelerated by frictionless, low-cost synthetic storytelling. Whether this campaign proves scalable — or remains a symbolic two-month sprint — depends on whether enforcement mechanisms (e.g., automated detection of 'face theft', audit trails for recommendation logic) are built into platform licensing requirements. As of November 6, 2026, no technical specifications for automated detection of 'face theft' or audit trails for recommendation logic have been published by NRTA.
Sources: IT之家，2026-11-06 "China's广电总局 launches AI micro-drama governance to curb low-quality synthetic content" — original report