Abstract:
Recently, the topic of "more than half of the country's computing power is creating cool dramas" has attracted widespread attention.
According to an "Investor Relations Activity Record Form" disclosed by A-share listed company Chinese Online, 55% of domestic AI Token consumption is burned in short dramas and video generation, 24% is burned in e-commerce advertising, and only 15% is actually used for research and development.
Token consumption generated by AI short dramas and videos has reached more than twice that of software development, becoming the largest core consumption scenario in the entire Token economics.
In April this year, the "AI-driven global micro-short drama innovation and development report" jointly completed and released by the Development Research Center of the State Administration of Radio and Television and others showed that in 2025, the global micro-short drama market will enter a period of explosive growth, with the global market size approaching US$18 billion, of which China's market size exceeded US$14 billion, accounting for 77.78% of the global market.
Other data shows that in the first eight months of this year, 430,000 micro-short dramas were launched in China, 13 times the number in the whole of last year.
On Douyin alone, 220,000 new AI short dramas were launched in the first half of the year. A 12-episode short drama would cost nearly 200,000 yuan in computing power.
Chinese Online Analysis believes that the core reason for the high consumption of AI Tokens is that short dramas are a business that requires extremely high video generation and Token turnover speed.
In terms of specific forms, AI simulation human dramas have stricter requirements on frame number and 1080p high definition, and their token consumption is usually two to three times that of AI comic dramas with a comic-like format and 720p resolution.
In addition, during the actual production process, in order to meet the director's strict requirements for plot narrative and shots, the team needs to continuously generate and screen images through "drawing cards", resulting in continuous and intuitive Token consumption.
Comments