Written by: Zheng Minfang
Source: Wall Street News
While OpenAI across the ocean seemed to have pressed the "pause button" on its AI generative model Sora, Chinese tech giants launched a counterattack in this field.
Recently, ByteDance launched its latest AI video generation model, Seedance 2.0, which quickly went viral on the internet thanks to its advantages such as multimodal input, self-camera operation, and consistency.
After experiencing the game extensively, Feng Ji, founder of Game Science, made a significant judgment: "The content field is bound to experience unprecedented inflation."
Feng Ji's prediction was not unfounded.
This shockwave is rapidly spreading to industries such as e-commerce, gaming, video platforms, and film production.
- In the e-commerce sector, the technical barriers to low-end outsourcing and filming bases have been completely eliminated;
- In the gaming industry, the production cycle for proof-of-concept and user acquisition materials has been compressed to the extreme, making competition even more brutal;
- Video platforms have had to further optimize their distribution logic to cope with the surge in supply;
- The traditional linear process of "shooting + editing" in film and television production is facing a devastating blow from the industrialized pipeline of "prompt words + generation".
A major industry reshuffle concerning beneficiaries and substitutes has already begun.
Explosion of video production capacity
Over the past year, the biggest pain point for AI video has been delivery.
Whether it's Sora, Runway, domestic platforms like Keling, or even ByteDance's self-developed JiMeng, they all suffer from this problem. Creators are often caught in a "gacha" game, needing to generate dozens of times to get a consistent, unbroken few-second video.
The core breakthrough of Seedance 2.0 lies in its attempt to transform "showy skills" into "deliverable narratives".
Breakthroughs in key capabilities are mainly reflected in three aspects:
- Firstly, it supports multimodal input . According to All-Weather Technology's actual testing, first-time users of Seedance can directly use Seedance 2.0 by activating automatic renewal for only 1 yuan. It supports text, images, videos, and audio as reference materials, meaning that almost any format you can think of can be input to generate videos.
- Secondly, it helps users understand the narrative and learn camera movement independently. Seedance 2.0 demonstrates "director-level" thinking, not only understanding complex narrative logic but also automatically directing camera language to complete push-pull, pan, and tilt camera movements. The video is no longer a simple movement of static images but possesses a cinematic narrative logic.
- Thirdly, there's the issue of image consistency . According to All-Weather Technology's tests on multiple AI video generation applications on the market, problems such as distorted facial expressions and blurred or intertwined backgrounds frequently occur during the subject's movement.
However, judging from the demo video, Seedance 2.0 maintains consistency in facial and visual information during the main movement process, making coherent storytelling possible.
This means that AI video generation is transforming from a toy into a tool. This ability to turn video generation into a standardized industrial pipeline makes "everyone can be a director" more than just an empty slogan, which will also significantly reduce the cost of video production.
Feng Ji used the term "inflation" to describe this transformation.
"The production cost of general videos will no longer be able to follow the traditional logic of the film and television industry, and will gradually approach the marginal cost of computing power. The content field will inevitably usher in unprecedented inflation, and traditional organizational structures and production processes will be completely restructured. I believe that anyone who has used it will quickly understand that this prediction is by no means unfounded," said Feng Ji.
First wave of impact
When the marginal cost of video production approaches zero, business models built on the existing cost structure will be the first to be affected.
E-commerce, gaming, video platforms, and film and television production are likely to be the first sectors to be affected.
The most direct impact was felt first in the e-commerce sector.
Product demonstration, scene depiction, and function explanation videos do not rely on complex artistic narratives, but rather on clear information delivery.
With the widespread adoption of Seedance 2.0, the barriers to video expression for businesses have been completely eliminated. Low-end video outsourcing companies and Taobao shooting bases that previously relied on "information gaps" and "technical barriers" for survival will face a harsh winter, and video production may shift from professional outsourcing services to businesses' own daily operations.
Compared to e-commerce, the impact of AI video generation models on games may be relatively limited, but a revolution has already quietly begun.
The costs of world-building, proof-of-concept, and paid user acquisition materials/videos are decreasing exponentially. More projects will be validated at earlier stages and eliminated at earlier stages as well.
An insider at a Beijing-based game company told All-Weather Technology that the company has already started small-scale testing of Seedance 2.0.
AI video generation models are also changing the distribution logic of video platforms.
For platforms like Douyin and Kuaishou, videos generated by models such as Seedance 2.0 have led to an explosion in content supply, forcing the platforms to completely shift their core competitiveness to the "screening and distribution" mechanism. For example, whoever's algorithm can more accurately extract the gems from the massive amount of AI-generated content, and whoever has higher commercial conversion efficiency, will be the winner.
In the film and television industry, Seedance 2.0's multi-camera storytelling capabilities may reshape the production process.
In the past, the creation of a film or television work often followed a strict linear industrial process: first, a massive amount of footage was shot, and then the editor would select, assemble, and construct the narrative logic in the post-production room.
However, in the logic of Seedance 2.0, this boundary is becoming blurred.
During the filming process, there is a possibility that the set design could be generated at low cost by an AI model ; the model itself has an understanding of camera movement and narrative rhythm, and the "editing" work is actually completed simultaneously at the moment the video is generated.
AI no longer simply spews out scattered footage, but directly delivers a "finished film" with a coherent temporal and spatial relationship.
This means that the time-consuming post-production editing stage in traditional film and television production is at risk of being "dimension-reduced" by algorithms.
The future creative workflow may no longer be "shooting + editing", but "cue words + generation". The editor's role will shift from "operator" to "instruction engineer" or "aesthetic gatekeeper".
Although the videos generated by Seedance 2.0 are not perfect, and there is still room for improvement in logical details and visuals, these challenges will not be obstacles in the near future given that the pace of technological iteration far exceeds market expectations.
The "moat" of IP
Seedance 2.0's amazing "reproduction" ability allows ordinary people to enjoy the pleasure of creation, while also putting unprecedented pressure on copyright holders.
Recently, a large number of "remixed" and even "parody" clips of Stephen Chow's classic movies have gone viral on short video platforms.
With the computing power of AI video generation models, Stephen Chow's facial expressions, signature laughter, and even classic dialogue style have been replicated by a large number of users at low cost, and even many absurd plots that have never happened before have been generated.
This quickly caught the attention of Stephen Chow's team.
Stephen Chow's agent, Chen Zhenyu, publicly questioned: "I would like to ask, does this constitute copyright infringement (especially given the large-scale dissemination these past two days)? I believe the creators have already profited, and is this platform simply allowing users to generate and publish these content without any oversight?"
This question, while seemingly revealing the copyright anxiety in the AI era, actually demonstrates the extreme scarcity of top-tier IPs in the AI era from a deeper business perspective.
In the future, amidst the deluge of AI-generated content, technology itself will no longer be a barrier, because everyone will have access to the same Seedance 2.0 tools.
The real barriers still lie with the IP owners.
It is precisely because the market is flooded with a large number of "high-quality imitations" of Stephen Chow that the irreplaceable "real Stephen Chow" IP is all the more evident.
When content supply is not only excessive but also "inflated," users' time and attention become more expensive than ever before. What can instantly capture users' attention are still those classic IPs that have stood the test of time and possess strong emotional resonance.
In other words, while AI lowers the barrier to entry for production, it infinitely increases the value of "recognition".
For IP owners, the future remains bright. IP assets accumulated over many years will no longer be merely objects of infringement, but will be able to achieve exponential growth in commercial value through official licensing, leveraging AI, and reaching countless creators.
From OpenAI's Sora 1.0, launched in February 2024, becoming the world's first AI video generation model to support the generation of 60-second videos, to ByteDance's Seedance 2.0, which now generates 60-second native audio narrative films from multimodal inputs, only two years have passed.
In this era of rapid technological development, all industries are standing at a crossroads: the cost of execution is being compressed infinitely, and those repetitive jobs that rely on manpower and long hours will be ruthlessly replaced; at the same time, the value of IP and creativity is being amplified infinitely.
When tools become readily available, what determines the quality of content will no longer be whether one knows how to use software, but whether the conception of the world in one's mind is unique enough.
