Displaced actors, stolen faces, and a seven-day regulatory blitz. China’s AI short-drama boom shows how quickly generative video can become a real industry.
"China has settled the first question for every entertainment market: AI production can reach industrial scale. The question that remains is who absorbs the cost when it does. The performers whose livelihoods dissolve. The individuals whose faces are harvested. The audiences asked to watch content that nobody invested enough to make compelling. Or the institutions that arrive to govern a transformation already well under way." IHistory repeats.
This is an insightful article. I am consistently impressed by the speed of China’s legislative processes. In an era of rapid technological disruption, the ability to be both robust and agile is a decisive strategic advantage. While no system is without flaws, Chinese institutions demonstrate a remarkable capacity for rapid, decisive action that reflects a strong connection to the societal needs they serve. History shows that those who define the rules set the standards, control the narrative, and ultimately lead the race. Conversely, entities resistant to change risk obsolescence. China’s less rigid historical framework regarding IP and its developing film industry have inadvertently provided leverage, allowing the country to adapt quickly and establish international standards in AI before competitors can react.
I'm doing something similar in Australia, but with a completely different angle. I use this to capture attention and promote our company, Ascendnce. Such content would not have been possible a few years ago due to the associated costs, but now it's within a startup's budget.
I posted the video on LinkedIn, and I was utterly slaughtered by both Australians and Americans, claiming it to be rubbish, inauthentic or low quality, that's nowhere close to a real production house.
What I found interesting is that, we don't aim to win an Oscar with this, nor do we think it's be anywhere close to a real production house. It's an experiment with a different strategy. Its main purpose is to capture attention and promote a business, which would not have been possible with real actors. We did put in a lot of effort in the script writing with someone whose profession is writing film scripts, incl the video creation itself. However, because it includes an AI element, some people dismiss it outright and even overlook its original idea and purpose.
This is a great example of exactly what the data shows at scale. In China, production costs dropped from $700 to as low as $27 per minute. The result: 470 new AI dramas per day, companies scaling from zero to 200 employees in a year.
Your use case is different from entertainment, but the economics are identical. When cost drops 90%+, entirely new categories of content become viable.
The audience reaction you got on LinkedIn is also consistent with what I found. In China, photorealistic AI content registered the lowest willingness to pay of any format. People detect the synthetic quality and resist engaging emotionally. The 1.4/10 audience score for an AI series in India that got 26.5 million views tells the same story.
My read: for attention capture and business promotion, AI video already works. For emotional connection, the gap is still real. Your framing of “we’re not trying to win an Oscar” is exactly the right way to think about it.
I have full confidence, that it will change overtime where people will pay and engage emotionally.
What’s used to create is merely a mean, what always matters is the output. And output itself will always have its own audiences, regardless how good or bad it is.
I reckon it was the same when they first invented and introduced photography. People at the time most likely suggested the exact same sentiment, where it is soulless and nowhere close to painters. I bet the photographer received the same critics. However, as we all have witnessed, photography becomes its own category and people even pay millions for it.
The photography analogy is fair. New mediums do eventually find their own audiences and value.
One thing the China data adds: 127,800 AI dramas in circulation, but only 0.117% crossed the 100-million-view threshold. Audiences are already sampling at massive scale. The gap is between sampling and emotional commitment.
My read is the technology will get there. The open question is how long that takes, and what happens to the people displaced during the transition.
I just watched your video. The issue isn’t using AI, but rather the quality of the content. Your video has very weird transitions, changing backgrounds, lips that are out of sync with the voices, unnatural audio, and a lack of artistic touch. You know that making a quality movie is largely about lighting, camera angles, music, acting, and color grading. If you don’t know the vocabulary or how to prompt the AI effectively, it won’t work magic. Using AI doesn’t remove the need for knowledge of filmmaking arts. In the future, using a less generic AI or one specialized in filmmaking might be better. I like your excitement and motivation. I fully support such endeavors, but you need to understand what makes art, art: it’s creativity and technique. Unfortunately, your video has neither yet, but don’t give up. Try again and keep improving. 加油
"China has settled the first question for every entertainment market: AI production can reach industrial scale. The question that remains is who absorbs the cost when it does. The performers whose livelihoods dissolve. The individuals whose faces are harvested. The audiences asked to watch content that nobody invested enough to make compelling. Or the institutions that arrive to govern a transformation already well under way." IHistory repeats.
This is an insightful article. I am consistently impressed by the speed of China’s legislative processes. In an era of rapid technological disruption, the ability to be both robust and agile is a decisive strategic advantage. While no system is without flaws, Chinese institutions demonstrate a remarkable capacity for rapid, decisive action that reflects a strong connection to the societal needs they serve. History shows that those who define the rules set the standards, control the narrative, and ultimately lead the race. Conversely, entities resistant to change risk obsolescence. China’s less rigid historical framework regarding IP and its developing film industry have inadvertently provided leverage, allowing the country to adapt quickly and establish international standards in AI before competitors can react.
I'm doing something similar in Australia, but with a completely different angle. I use this to capture attention and promote our company, Ascendnce. Such content would not have been possible a few years ago due to the associated costs, but now it's within a startup's budget.
I posted the video on LinkedIn, and I was utterly slaughtered by both Australians and Americans, claiming it to be rubbish, inauthentic or low quality, that's nowhere close to a real production house.
What I found interesting is that, we don't aim to win an Oscar with this, nor do we think it's be anywhere close to a real production house. It's an experiment with a different strategy. Its main purpose is to capture attention and promote a business, which would not have been possible with real actors. We did put in a lot of effort in the script writing with someone whose profession is writing film scripts, incl the video creation itself. However, because it includes an AI element, some people dismiss it outright and even overlook its original idea and purpose.
You can have a look at the responses here: https://www.linkedin.com/posts/jpctan_ai-genai-solofounder-activity-7435866979019935745-MygE
You know what? I've commissioned a 10-episode series, which I'm committed to both the writer and creator. I'm going to keep producing it.
This is a great example of exactly what the data shows at scale. In China, production costs dropped from $700 to as low as $27 per minute. The result: 470 new AI dramas per day, companies scaling from zero to 200 employees in a year.
Your use case is different from entertainment, but the economics are identical. When cost drops 90%+, entirely new categories of content become viable.
The audience reaction you got on LinkedIn is also consistent with what I found. In China, photorealistic AI content registered the lowest willingness to pay of any format. People detect the synthetic quality and resist engaging emotionally. The 1.4/10 audience score for an AI series in India that got 26.5 million views tells the same story.
My read: for attention capture and business promotion, AI video already works. For emotional connection, the gap is still real. Your framing of “we’re not trying to win an Oscar” is exactly the right way to think about it.
I have full confidence, that it will change overtime where people will pay and engage emotionally.
What’s used to create is merely a mean, what always matters is the output. And output itself will always have its own audiences, regardless how good or bad it is.
I reckon it was the same when they first invented and introduced photography. People at the time most likely suggested the exact same sentiment, where it is soulless and nowhere close to painters. I bet the photographer received the same critics. However, as we all have witnessed, photography becomes its own category and people even pay millions for it.
The photography analogy is fair. New mediums do eventually find their own audiences and value.
One thing the China data adds: 127,800 AI dramas in circulation, but only 0.117% crossed the 100-million-view threshold. Audiences are already sampling at massive scale. The gap is between sampling and emotional commitment.
My read is the technology will get there. The open question is how long that takes, and what happens to the people displaced during the transition.
I just watched your video. The issue isn’t using AI, but rather the quality of the content. Your video has very weird transitions, changing backgrounds, lips that are out of sync with the voices, unnatural audio, and a lack of artistic touch. You know that making a quality movie is largely about lighting, camera angles, music, acting, and color grading. If you don’t know the vocabulary or how to prompt the AI effectively, it won’t work magic. Using AI doesn’t remove the need for knowledge of filmmaking arts. In the future, using a less generic AI or one specialized in filmmaking might be better. I like your excitement and motivation. I fully support such endeavors, but you need to understand what makes art, art: it’s creativity and technique. Unfortunately, your video has neither yet, but don’t give up. Try again and keep improving. 加油
Thank you for your feedback. Do you want to join us and help create the series 2?