The videos were generated by Seedance 2.0, an artificial intelligence tool developed by ByteDance, the parent company of TikTok.
With a short line of text, users could summon short films complete with soundtracks, camera movement and dialogue — often featuring copyrighted characters. Within days, major studios were drafting cease-and-desist letters.
It was not the first time generative video technology had unsettled the film industry. Tools such as Sora, Runway’s Gen-3 Alpha and Pika had already previewed a future in which movies could be produced from prose.
But Seedance struck a deeper nerve — not only for its technical quality, but for how easily it collapsed the boundary between fandom, piracy and production.
When Seedance-generated clips featuring Marvel and Pixar characters went viral, Disney and Paramount accused ByteDance of facilitating copyright infringement. Japan’s Agency for Cultural Affairs soon opened an inquiry after anime-style shorts appeared online without authorization.
ByteDance said it was strengthening safeguards.
In 2023, The New York Times sued OpenAI and Microsoft over training data. A year later, Reddit filed suit against Perplexity over unauthorized scraping.
Together, the cases expose a growing contradiction: media companies condemn unlicensed data use, even as they quietly integrate AI into their own production pipelines.
Disney, according to industry sources, has since signed a licensing agreement with OpenAI’s Sora worth roughly $1 billion.
Beneath the legal disputes lies a more unsettling question: Is artificial intelligence creating anything at all?
“If a user is required to initiate the process, then the AI has not really created the film,” said Gwen Grewal, a philosopher at The New School for Social Research. “It has compiled it. Like a production manager or editor.”
Without consciousness, she argues, there is no intention — and without intention, no true authorship.
Grewal traces the issue to existential philosophy, from Heidegger to Sartre, which links creativity to human awareness of limits.
“People are born into systems they did not choose,” she said. “But they know they can become something more. Does AI have that impulse? Can it reject its own work? Can it care if it fails?”
Seedance can generate surprises, even for its engineers. But it cannot “mean” what it produces.
That distinction, she suggests, may define the boundary between creation and computation.
Not everyone in the film world sees AI as an existential threat.
“I don’t think directors have much to fear,” said Shin Haerin, a professor at Korea University’s College of Media and Communication.
“What AI cannot replicate is intention,” she said. “That will become more valuable, not less.”
She compares the shift to fashion.
“As we distinguish between ready-to-wear and haute couture, cinema will divide between automated content and intentional filmmaking.”
In this view, AI will dominate fast, disposable entertainment — short clips, personalized stories, viral visuals — while human filmmakers retreat into slower, more deliberate work.
Polish will matter less than purpose.
That Seedance emerged from Beijing is no accident.
Over the past year, China has placed generative AI and robotics at the center of national strategy, investing heavily in chips, automation and algorithms.
The rise of Seedance follows the breakthrough of DeepSeek, which became the most-downloaded free app in the U.S. Apple Store in 2025, briefly overtaking ChatGPT.
Together, they signal China’s ambition to compete not just in manufacturing, but in cultural technology.
For Beijing, AI is also a soft-power tool.
By enabling users to generate cinematic content in Mandarin, Cantonese and minority languages, platforms like Seedance could reduce reliance on Hollywood imports and strengthen domestic entertainment ecosystems.
What begins as hobbyist creativity may evolve into a parallel film industry.
Seedance 2.0 has not ended cinema. Nor has it solved creativity.
But it has exposed how fragile the old boundaries have become — between studio and fan, artist and algorithm, inspiration and imitation.
In the coming years, the central question may no longer be whether AI can make films. It will be whether audiences still care who made them.
Copyright ⓒ Aju Press All rights reserved.