AI-Generated Video Is Reshaping China’s Entertainment Workforce

China’s short-drama market is already seeing major AI substitution. In Q1 2026, about 128,000 short dramas were published in China, three times the total for all of 2025, and 95 percent were AI-generated, according to the China Netcasting Services Association cited in the article.

The shift matters because the format rewards speed, volume, and low production costs. Tools such as ByteDance’s Seedance 2.0 have made digital performers capable of producing videos faster and cheaper than traditional shoots, according to the report.

The article cites Tsinghua University professor Shen Yang telling the Financial Times that one minute of AI video now costs about $90 to $120, around ten percent of what production with human actors used to cost. That price difference gives producers a strong incentive to replace human performers in high-volume content.

The impact reaches beyond actors. The report says China’s entertainment industry directly employs 690,000 people, while 15 million list livestreaming as their primary job, making automation a workforce issue as much as a production trend.

Some performers are reportedly being forced to “distil” their voice and likeness into AI tools before being fired. That creates a direct conflict over whether a performer’s identity can continue generating content after their role ends.

For workers, the key takeaway is to scrutinize contracts involving voice, image, likeness, training data, and post-employment usage. For studios and platforms, the risk is that unclear consent practices can turn cost-saving AI adoption into legal and reputational exposure.

Lawyers cited in the article say AI-related labor disputes have risen over the past two to three years. The timing aligns with rapid improvements in AI video systems and wider use of digital performers in entertainment production.

The report also notes that newer systems such as Seedance 2.5 and Alibaba’s Wan 3.0 have pushed AI video quality higher again. As output becomes more realistic and commercially usable, the central question is no longer whether AI video can be used, but how rights, pay, and consent will be handled.