---
title: "Cutting Per-Episode Costs by Two-Thirds and Multiplying Capacity Tenfold: AI Motion Comics Are Rewriting the Content Industry"
author: "Unique Research"
sourcePublication: "Unique Research Substack"
originalPublishedAt: "2025-12-16T08:01:25+00:00"
canonical: "https://ffcap.cn/en/research/src-20251216-01html"
source: "https://uniqueresearch.substack.com/p/src-20251216-01html"
language: "en"
---

# Cutting Per-Episode Costs by Two-Thirds and Multiplying Capacity Tenfold: AI Motion Comics Are Rewriting the Content Industry

_Original · Unique Research · 2025-12-16_

_Historical edition note: This complete December 16, 2025 translation preserves the full six-part analysis and all four conversation sessions. Production costs, output, revenue, audience figures, rankings, forecasts and industry-wide claims remain attributed to the speakers or original report, not current or independently audited measurements. Currency is not specified for the monetary figures in the source. The headline's two-thirds cost reduction and tenfold capacity wording is retained as the original framing, not a separately established calculation. Mantalk (漫谭), Yijie Creative (艺界创想), Jellyfish Intelligence (水母智能) and translated project titles are source-based English renderings, not claims of verified official English names. The source itself labels the incomplete title “捡妖...” and the Luma/Sora/DiT passage as conjectural transcriptions; those uncertainties remain and the latter does not establish who originated the DiT architecture._

If you have recently watched short videos, you may have noticed a small change: beside live-action wish-fulfillment dramas that simply cut abruptly, many moving comics have suddenly appeared.

The characters are two-dimensional, the shots cinematic, and the pacing belongs to short video. Someone in the comments asked: is this not an animation-industry revolution standing on short video's shoulders?

At a super-individual workshop during the 2025 Beijing UniqueBloom, moderator Zhao Liang, partner at Unique Capital, and three guests—Mantalk producer Mai Yun, Yijie Creative founder Chai Wen, and Jellyfish Intelligence co-founder Zhou Zhipeng—clarified the sober logic behind this revolution: AI motion comics involve far more than the four Chinese characters for image-to-video.

I. The Explosion of AI Motion Comics Was No Accident, but the Inevitable Result of Three Forces Converging

Why did AI comic dramas move from "industry experimentation" to an "internet-wide buzzword" within a few short months?

Mai Yun's answer was direct: technology providers, platforms, and producers all found the same rhythm at the same moment.

First, technology suddenly accelerated.

Traditional animation, whether 2D or 3D, once took months or years, with more than ten roles arranged in a long chain: original art, storyboards, coloring, compositing, and more. Every stage consumed money and time.

Now AI cuts that chain in half. Many intermediate roles contract into a small director + editor + AI team, which can complete a series in 30–40 days at a cost that has moved from unimaginable to just affordable enough to try.

Then platforms smelled the opportunity.

Douyin, Kuaishou, Bilibili, and iQIYI launched comic-drama support programs almost simultaneously over the preceding months because they saw clearly that short video's dividend had peaked and live-action short dramas were brutally competitive. New growth required opening a new content category.

AI comic dramas perfectly satisfy three criteria: scalable production, a fresh form, and a replicable business model.

Then creators' attitudes changed.

In the traditional industry, animation is highly specialized and extremely unfriendly to individual creators. AI removed a large share of the barrier:

You cannot draw but have aesthetic judgment—that works;

You lack formal training but have watched enough films—that also works.

Large numbers of people who could never have entered animation were drawn into the field.

Chai Wen summarizes it as production-side efficiency colliding with consumer-side demand.

When technology becomes mature enough to land in a field, an explosion is only a matter of time.

Zhou Zhipeng added that content forms themselves are changing. Comic platforms' DAU has declined markedly over the past two years, while long dramas and films are also falling. Short video and fragmented content are growing wildly.

Live-action short dramas have competed for five years and serve female users well, but male-oriented themes and more vertical content remain absent. AI comic dramas fill the gap: they control costs while supporting every kind of subject, suddenly electrifying the market.

Technology, platforms, creators, and user habits all arrived together. That full hand of four cards—not the accidental success of one breakout tool—created today's AI motion-comic boom.

II. AI Makes Materials; Humans Make Works

The technological climax arrived quickly, but an old truth still determines the industry's direction: technology is only a tool.

Mai Yun recalled their disappointment upon first encountering AI in 2024: rough visuals, broken faces, and no basic consistency, far from anything that could be called a work.

As model capabilities iterated repeatedly in 2025, one day people suddenly realized that AI could now complete a watchable story and even convey some artistic expression.

Disappointment became excitement, followed by forced calm: what exactly can AI help me do? Which stages can I entrust to it, and which must I supervise personally?

Chai Wen's perspective is representative. She comes from traditional film and worked on projects such as The Battle at Lake Changjin and The Volunteers, giving her deep familiarity with industrial processes. She says AI output can currently be called materials at most, not works. What elevates materials into a work is years of project experience, a director's viewpoint and narrative rhythm, and integrated judgment about emotion, cinematic language, and aesthetics.

In other words, AI can accelerate the visualization of imagination, but cannot replace understanding of humanity, emotion, and values.

AI primarily assists work and output, efficiently integrating and expanding boundaries according to creators' accumulated knowledge, depth of thought, experience, and understanding; creators ultimately determine the finished content. The source writes Al, evidently referring to AI.

Zhou Zhipeng uses his company's history to identify four stages in AI image/video technology:

Stage one: people played with models on Discord and technology was a toy;

Stage two: the "crossing the dimensional rift" wave ignited creative desire;

Stage three: LoRA and ControlNet emerged, evolving AI from a toy into a productivity tool;

Stage four: from the second half of 2023, technology truly embedded itself in specific industries such as comic dramas.

To him, today's stage resembles the early years of mobile internet: silly applications everywhere, short life cycles, yet tremendous vitality.

You can laugh at the roughness, but cannot ignore the possibility space being unlocked.

This is today's contradiction: astonishing efficiency beside still-rough works; excitement over an unseen ceiling beside anxiety that foundational aesthetics have yet to emerge.

The real answer probably lies in the next section—how people and AI jointly bring an AI comic drama into being.

III. How Is an AI Motion Comic Incubated?

Treat an AI comic drama as a production line and two pipelines spiral forward together: comics and image-to-video/multimodal generation.

All intelligence serves one core question: how can key roles—topic selection, screenwriting, directing, and storyboarding—maintain professional judgment under high concurrency?

At Jellyfish Intelligence, the process begins with novel IP.

They have signed more than 300 novels upstream and hold a regular monthly selection meeting. A selection Agent first performs a rough screening of genre, pacing, character relationships, density of satisfying moments, and user profiles; senior selection specialists then make a second judgment.

AI supplies the data perspective; humans hold the intuitive perspective.

Once a topic is chosen, adaptation begins.

For a long-running comic, AI can adapt 80% of the content, breaking text into storyboards, dialogue, and pacing. For a two-minute comic drama, however, pacing is extremely tight and emotions must rise and fall every dozen seconds. AI can currently handle only 20%–30%; screenwriters must deeply participate in deleting and revising plot, intensifying conflict, and adjusting rhythm.

Next come visuals and generation.

AI involvement is highest here. Platform users are highly diverse: high-school students, working civil servants, and women in their fifties all use the tools to create.

You can think of AI as abstracting drawing into parameters and templates, enabling people without hand-drawing skills to participate in visual production.

Several stages remain beyond AI's reach—or far below deliverable quality even when touched:

Rough storyboards concern cinematic language, not whether a single frame is attractive. AI can generate a first directing storyboard, but the director revises it and determines pacing and emotional direction. High-quality comic dramas still rely on human voice actors because AI cannot yet match the emotional ceiling and rhythmic delivery.

Once these responsibilities are clear, the cost structure changes completely.

Zhou Zhipeng offered direct figures: their earliest comics cost more than 7,000 (currency unspecified) per episode; today, a premium AI comic-drama episode can be produced for 1,000 or 2,000 (currency unspecified).

When costs move from permitting only a dozen or so works to allowing 50 experiments, the industry's method of experimentation changes too.

The question is no longer whether one project will succeed, but what can succeed across a broad portfolio.

IV. From PUGC to UGC: Whose Foundation Will Institutions Become?

Once the production line is redefined, another question surfaces: what role remains for institutions in a future ecosystem potentially dominated by UGC?

An audience member asked directly whether institutions would be marginalized if a full-lifecycle UGC ecosystem emerged.

Zhou Zhipeng answered clearly: the future looks like a flat network, not a pyramid topped by giant smokestack factories.

Platforms will still control traffic gateways in the short term. Institutions shift from workshops to infrastructure: providing capabilities by turning years of experience into creator-friendly tools and templates; providing knowledge bases that preserve cases, lessons, and successful paths for countless creators; and supporting collaboration so geographically dispersed creators can maintain unified styles, consistent characters, and controllable output quality.

From traditional film, Chai Wen reaches a similar conclusion: AI removes many technical barriers, but institutions must still integrate resources, accumulate experience, and operate brands over the long term.

She prefers to view her team as a core experience node that can connect more co-creators and repeatedly produce high-quality content.

Mai Yun opens the door further: even if you cannot draw, you can enter if you embrace AI and creation and possess aesthetic judgment and sufficient viewing experience.

He also warns that directing and editing may look easy to learn and replace, but ordinary people remain far behind genuinely excellent directors.

In other words, AI greatly flattens horizontal skill differences, but the vertical heights of top experts remain.

Institutions will not disappear; they will retreat from centers of power into foundations supporting every super individual.

V. Will AI Comic Dramas Produce the Next Journey to the West?

An audience member asked the event's most romantic and practical question: will this field produce a universal hit like Journey to the West or Jin Yong's novels, carrying a generation's memories?

Mai Yun is cautiously optimistic. At a broader level, he believes universal hits should be judged from the perspective of AI film rather than only the comic-drama subtype.

AI 2D technology is relatively mature, but producing a nationwide phenomenon at the level of Ne Zha will still take several years.

He can say only that there is hope; the concrete form remains invisible.

Chai Wen is more decisive: a hit is inevitable, but content will separate into three tiers.

At the bottom is massive fast-consumption content satisfying broad entertainment demand;

In the middle is high-quality vertical content that gradually accumulates loyal, engaged audiences and builds brand influence;

At the top are a very small number of timeless masterpieces that are not merely entertaining but become a generation's memories and accompany their shared growth, like Journey to the West.

These tiers cannot emerge from the same mechanism or be completed by the same people. They require different ecosystems: some participants experiment rapidly, some refine work over the long term, and some push it to the limit.

Zhou Zhipeng pours on some cold water: do not expect the next Journey to the West yet.

For one thing, China's anime-culture foundation differs entirely from Japan and Korea and remains relatively niche;

For another, today's AI comic dramas primarily target wish-fulfillment-drama audiences seeking fast-consumption content.

At this stage, comic dramas' value remains largely short-term commercial returns and can hardly support lasting IP value.

His judgment is that epoch-making IP can emerge only after AI reaches blockbuster-film capability, when it can not only create sufficiently strong visuals but support highly complex narratives and character development.

Today's AI motion comics are therefore more like an outpost on the road to the future: probing new content structures while using countless small works to train a larger, deeper narrative engine.

VI. Why Is Now the Best Time to Enter AI Motion Comics?

For institutions, the message is to stop imagining another giant factory and become new infrastructure that productizes capabilities and powers tens of thousands of creators.

For creators—screenwriters, directors, comic artists, or individuals just learning AI tools—the gift of this era is that technique is rapidly being leveled; what differentiates you instead is your understanding of stories, people, and the era.

Knowing LoRA is no longer the entry barrier. Thinking clearly about what you want to express is the new dividing line.

For platforms, the real question may not be how many series to support, but how to design more refined traffic mechanisms so people with genuine desire to express and sustained ability to iterate are seen, rather than simply pouring traffic onto several merchandise-selling wish-fulfillment dramas.

For ordinary viewers, today you can almost move from being fed content to writing a little yourself.

You need not make a feature film; a 30-second story is enough to leave a mark in this ecosystem.

In this sense, the real upgrade brought by AI motion comics is not resolution rising from 1080P to 4K, but an upgrade in the clarity of expression.

It moves from having many feelings but being unable to articulate them to finally transferring the images in your mind onto the screen, piece by piece.

Technology will continue iterating, tools will update generation after generation, platform policies will change, and trends will shift.

One thing remains relatively stable: humanity's desire for stories has never diminished.

AI will not tell your story for you; it is quietly dismantling the overly thick wall between you and the story.

If you already have an image, a line of dialogue, or a character in mind,

Today may be the best starting point for entering the world of AI motion comics.

More Details from the Conversation

Session One: Guest Introductions and Product Demonstrations

Zhao Liang: I am Zhao Liang, partner at Unique Capital. Today's Panel may differ slightly from the others because our themes are "comic drama" and "motion comics." I believe we need multimedia demonstrations to give everyone a more concrete impression.

In the first session, each of our three guests will spend 5 to 10 minutes demonstrating their company's products and services. The second is a roundtable lasting roughly half an hour to 40 minutes. In the third, mystery guests and audience members will ask the three guests questions.

Let us begin with part one. Please welcome Mai Yun from Mantalk to introduce Mantalk and its products. I will first play a short video.

Mai Yun: What you just saw resembled a promotional PV and also traces Mantalk's journey. At the beginning, The Slit-Mouthed Woman was still a traditionally produced AE motion comic. Later, Picking Up Demons...—presumably a work title—and The Zoo had fully shifted to production with AI comic-drama tools.

Specifically, I feel the video gradually becomes more spectacular and attractive from beginning to end. If the audience were children, they might already be thrilled. That reflects our projects. Mantalk integrates production and distribution. This PV does not include our October projects Infinite Returns or Fighting Zombies, both of which have strong data. We hope everyone will support them.

Zhao Liang: Thank you, Mai Yun. Next, Chai Wen, founder of Yijie Creative.

Chai Wen: Hello, everyone. I am Chai Wen, founder of Yijie Creative.

My industry background is somewhat complex. I previously worked as a product manager at Baidu and Opera Browser, then founded a film-production company. Many people say those fields are unrelated, but I believe the work is the same: we create and design a product, and the people we serve are users.

Our company now has complete visual-effects and animation production teams. We have been fortunate to participate in well-known projects including The Battle at Lake Changjin, The Volunteers, The Wandering Earth, and Moon Man. Drawing on that accumulated experience, we now primarily build creative teams to incubate our own IP content.

Going forward, we may combine our experience and technology, industry understanding, and AI tools to produce better content that serves users. That is our current structure, and AI assistance has been integrated into every part. Our prior industry background allowed us to encounter AI technology relatively early.

We now mainly experiment with short dramas and animation because, whether in model performance or feasibility, AI accelerates the path from idea to output in both fields. We will continue exploring other possibilities.

Zhao Liang: Finally, Mr. Zhou from Jellyfish Intelligence.

Zhou Zhipeng: Hello, everyone. I am Zhou Zhipeng and am delighted to be here. The previous two companies are impressive. We now cooperate with Mantalk and entrust our works to its highly professional distribution.

I also have a Beijing company called LKK Innovation Design Group that has operated for 20 years. We later opened a new company in Hangzhou, today's Jellyfish Intelligence. Chushou AI is a product platform under Jellyfish, an AI animation-creation platform.

Our principal domestic business is producing and distributing comic dramas. The posters show some current series with strong data. This year, we have roughly 24 series running online and may have cumulatively launched 50 by the end of December. Among the better-performing series, the second has generated more than 3 million in revenue (currency unspecified) and approximately 300 million views (the source combines “nearly” with “more than” in this figure). Almost every work begins as a comic and also appears as a comic drama.

In Hangzhou we focus on product research, with algorithm scientists and product teams jointly building Chushou. Over the past two or three years, since beginning foundation-model work in 2022, we have accumulated more than 3 million users and signed several hundred creators for deep training and joint commercial delivery.

Our commercial chain works roughly as follows: upstream, we acquire novels from novel-IP owners and now hold more than 300, with regular monthly selection meetings. After selection, we adapt them into comics and comic dramas. Our own team and signed platform creators jointly produce the comic dramas, which we then distribute downstream. We distribute domestically and overseas; domestically, we have around 1.4 billion views and 100 billion seconds of watch time.

Overseas, especially in Japan, we believe we were the first to produce AI comics and receive a strong response. We cooperate well with Japanese platforms including PICCOMA, DMM, Unireel, and Webnovel. This AI wave creates a major opportunity for Chinese content to flow back into animation powerhouses such as Japan and Korea.

In our production chain, blue represents the current AI process and red the traditional industry. AI now empowers nearly every stage: a selection Agent for topic selection, an adaptation Agent matrix for adaptation, followed by character generation and training, visuals, animation, and more. Several stages remain entirely unreplaced by AI, including rough sketches, storyboards, cinematic language, and later voice acting, which remains human because emotional expression is stronger.

Our team is not purely technical but combines interdisciplinary talent: algorithm specialists, designers, animation editors, and directors. Expert knowledge greatly supports product iteration and evolution.

Overseas, we operate animeshorts.ai. Unlike our domestic platform, it uses robot conversations to help users create comics and light comic dramas and is primarily UGC. Japan has two kinds of deeply engaged users: boys' love fan creators, a highly popular category there, and daily educational creators, such as a 60-year-old man who draws more than ten pieces each day about science and his life.

The flywheel works as follows: data from Douyin and other platforms—including per-episode completion rate, 30-second completion rate, retention, and comments—returns to train our product, strengthen its "expert capability," and apply those capabilities to works. When works perform well, we package the experience and return it to UGC users, allowing ordinary people to create good work with a low barrier.

Although we currently produce PUGC (professionally generated user content), working alongside users, we believe UGC may ultimately dominate. Even IP building will no longer be one-way output; users will build with you, everyone participating and creating emotional connections. Thank you.

Session Two: Roundtable Discussion

Zhao Liang: Thank you, Mr. Zhou. Please take your seats for the second session and a discussion of industry questions.

Recently everyone seems to be discussing "AI comic dramas," from platforms and tool companies to producers. Especially since October, everyone has rushed into the field.

First question: how do you view this phenomenon? Why have AI comic dramas become so popular within a few months?

Mai Yun: Why did AI comic dramas suddenly develop so quickly? I believe it results from mutually reinforcing technology providers, platforms, and producers.

Technology development is the foundation. Traditional 2D and 3D animation took months or years. With AI, a series can be completed in as little as 30 or 40 days. AI not only shortens the cycle, but sharply reduces costs. Traditional animation may require more than ten roles; with AI, a director, editor, and AI technical team can complete an entire series.

Second are platforms. They saw the opportunity and, with content producible at scale, Douyin, Kuaishou, Bilibili, and iQIYI introduced their own "comic-drama policies" over the preceding months. Platforms want to capture incremental markets, discover users, and retain market share.

Finally come producers. We see advantages in cost and cycle and consider this a major opportunity and future trend. Joint reinforcement from all three drove the rapid development of recent months.

Zhao Liang: That sounds like perfect timing, location, and people. Ms. Chai, please share your view.

Chai Wen: I think production-side efficiency has met consumer-side demand. It is inevitable, not accidental.

The technology existed for a long time, but can be used now because it has become viable in short dramas and comic dramas. People watched for a long time; experimentation used to be expensive, while AI now lowers professional and entry barriers.

For platforms, this captures a strategic high ground in the next generation of content. Long dramas and major films once absorbed large investments; now the same time and money can fund hundreds of works, an order-of-magnitude difference.

For creators, the entry barrier is lower and more people dare to express themselves.

For users, there is an opportunity to see many people's expression rather than only that of a small group.

In technological iteration, creator feedback also provides AI with more experimental data, allowing accumulated data to improve future updates.

Zhou Zhipeng: I would add three factors:

Changing content forms: comic platforms' DAU has fallen sharply over the past two years, while long dramas and films are also declining. The rise of short-video fragments inevitably drives new content.

Short-drama bottlenecks: live-action short dramas developed for five years but growth weakened, with predominantly female users. Male users and more suitable themes urgently need to be brought in. Comic dramas offer a boost by solving cost and subject limitations, becoming a breakthrough.

Capacity breakthrough: advanced animation and comic capacity was once extremely limited and could not support high concurrency. AI comic dramas provide at scale, and the market already contains more than 10,000 such dramas, allowing the field to operate.

Zhao Liang: Second question: amid today's market noise, how do you feel—excited, anxious, or afraid?

Mai Yun: When we first encountered AI in 2024, we felt discouraged. The output was unwatchable and could not even maintain facial consistency.

In 2025, technology developed rapidly and our discouragement became excitement. We have not yet seen AI's upper limit, but can already tell desired stories and even express art. Many traditional-animation directors actively embrace AI: camera movement and composition that once took one week may appear in two minutes with the right keywords.

After the excitement, however, we believe we must remain calm and objectively analyze what help AI can provide.

Zhao Liang: You feel excitement today, but reason tells you to remain calm. Ms. Chai?

Chai Wen: I am highly energetic; when AI arrived, I primarily saw opportunity.

I consider how to integrate it into my workflow. Practice leaves me with this conclusion: AI output can currently be called only "materials." Our past experience, directing mindset, and artistic aesthetics turn it into a "work." Artificial intelligence cannot replace human warmth and emotional needs.

The future is coexistence with AI, treating it as an important partner that shortens the path from imagination to visualization.

Zhou Zhipeng: We experienced four cycles:

The first came when AI entered China and everyone played on overseas Discord;

The second was the "crossing the dimensional rift" wave on Douyin in November 2022, which stimulated creative desire;

The third was the rise of LoRA and ControlNet in March 2023, when AI became a productivity tool;

The fourth began in the second half of 2023, when it entered vertical industries.

When we proposed comic dramas in early 2024, no one understood; they exploded on Douyin by year-end. Today's stage resembles the first year of mobile internet, with massive numbers of APP products emerging. Many are silly and short-lived, but highly vigorous.

We expect premium works to display stronger commercial monetization next year around March–April or May–June. We are highly optimistic and consider this the inaugural year of comic drama. We feel extremely excited.

Zhao Liang: Yes, very excited. As people mentioned, after models based on the DiT architecture proposed by Luma and Sora launched—the original text here reads "from 11 sorry to ..." and is unclear, but appears to reference Luma/Sora/DiT—companies including ShengShu Technology with Vidu, Aishi Technology with PixVerse, Zhipu AI with CogVideo, and Runway have all been launching new tools rapidly. This year may truly be the inaugural year.

Session Three: In-Depth Q&A

Zhao Liang: A question for Mai Yun. Mantalk ranked first by Douyin comic-drama views in October by a large margin. How did you achieve that so quickly—content, tools, or paid traffic?

Mai Yun: Although we have produced comic dramas for only a short time, we have worked in content for years. We began with novel copyrights and produced audio comics and live-action dramas, establishing a solid content foundation. We simply happened to emerge at this moment in comic drama.

Most important is creating from the user's perspective: we make what users enjoy.

On production, we continually explore the most advanced technology.

Finally, we practice open collaboration. Across technology, distribution, and production, we improve together and jointly create excellent works.

Zhao Liang: A question for Mr. Zhou. Can you break down the AI comic-drama production process? How do people and machines collaborate?

Zhou Zhipeng: Our process has two pipelines rising in an intertwined spiral: comics and image-to-video/multimodal generation.

The core is increasing concurrency in key roles. AI cannot replace the professional understanding of directors, screenwriters, and selection specialists; we need to improve their efficiency.

Topic selection: a selection Agent performs the first evaluation, followed by a second human review.

Adaptation: AI can smoothly adapt 80% of a long work; in a two-minute comic drama with dense emotional beats, it can currently handle only 20%-30%, requiring deep screenwriter participation.

Visuals and generation: AI involvement is high. High-school students, civil servants, and women in their fifties all participate on our platform.

Irreplaceable stages: rough comic storyboards conveying cinematic language, director storyboards generated by AI and corrected by the director, and voice acting, where AI's emotional ceiling remains insufficient and premium comic dramas still require people. In cost terms, early comics cost more than 7,000 (currency unspecified) per episode; a premium comic drama now costs roughly 1,000 or 2,000 (currency unspecified).

Zhao Liang: Is the industry still chaotic? Some charge 20,000 per minute (currency unspecified), while others charge less than 1,000.

Another question concerns collaboration. Many producers subcontract work. Is that a temporary necessity or a future trend?

Zhou Zhipeng: We believe centralized capacity—hundreds of people sitting together as cogs—belongs to the past.

AI solves two problems:

Experience leveling: AI rapidly levels techniques that once required five years of accumulation, letting everyone output from the same starting line.

Controllable remote collaboration quality: with five partners dispersed across the country, we can achieve nearly identical output quality and character consistency. UGC will flourish in the future, especially interactive plots and multibranch content better suited to individual creators.

Zhao Liang: A question for Ms. Chai. As representatives of traditional film, what advantages and disadvantages do you have entering AI comic drama? Will you create premium work or silly dramas?

Chai Wen: Our strategy is clear: we are not an AI technology company, but a content-creation company.

First, we ensure we do not fall behind; then we continually explore advanced technology and use core strengths accumulated over years—production experience, narrative understanding, cinematic language, and visual-style control—to produce good works continuously. We are not limited to comic dramas and will enter live action as models mature.

We also maintain an Open mindset, package experience as products, attract more excellent creators, and achieve co-creation and mutual success.

Session Four: Audience Interaction

Audience Member 1: If a full-lifecycle UGC ecosystem truly emerges, what role should institutions play?

Zhou Zhipeng: The future is a flat world with a networked distribution.

Institutions become foundational support:

Platforms provide traffic, which always remains concentrated.

Institutions provide capabilities, knowledge bases, and collaboration support, becoming infrastructure that serves every super individual. Content forms will also change; interactive plots, like watching a drama while playing a game, may become a major commercial opportunity.

Chai Wen: AI solves technology; institutions solve resources and experience. We hope to establish core barriers, include more co-creators, produce good content, and build brand influence.

Mai Yun: We also welcome individual authors. Anyone willing to embrace AI and creation can participate, even without drawing skills, provided they have aesthetics and viewing experience. Directing and editing may look easy to learn, but their ceilings are high; ordinary people cannot approach excellent directors.

Audience Member 2: Will AI comic dramas create a generation-defining universal hit like Journey to the West or Jin Yong's novels, or become more diverse and fragmented?

Mai Yun: A universal hit should not be limited to the term "comic drama" but viewed from the perspective of AI film.

Current AI 2D technology is relatively mature, but reaching nationwide-hit level, like Ne Zha, will require several more years. I can cautiously say there is hope, but it is not yet visible.

Chai Wen: Hits are inevitable.

Future content will have three tiers:

Basic mass-market fast-consumption products produced at scale.

Vertical content with brand influence.

At the top, hits and timeless masterpieces that influence a generation's thought and culture. These will necessarily emerge from different ecosystems.

Zhou Zhipeng: Anime culture is itself niche in China, unlike Japan and Korea. A Chinese hit must have Top-level production.

Comic dramas currently target wish-fulfillment audiences and are short-term fast-consumption products without lasting derivative IP value.

Only when AI reaches blockbuster film-production capability can true IP value and absolute hits emerge.

Zhao Liang: Thank you again to all three guests. I believe AI comic dramas will grow increasingly vigorous and attract many voices, but this is a good beginning. Thank you all.

---

Original publication: https://uniqueresearch.substack.com/p/src-20251216-01html
On-site reading page: https://ffcap.cn/en/research/src-20251216-01html