Everyone is caught up in the absurdity, paying tribute to the craftsmanship behind 'Niu Lai' for insisting on hand-crafting a Flash-style animated film in an era dominated by AI. I, on the other hand, am genuinely curious: how much would it actually cost to make a film like 'Niu Lai' using AI? And would it even be more cost-effective?

It might be inappropriate to ask industry insiders, as the conversation could easily hit a nerve. After all, hand-crafting an 86-minute animated film that makes it to the big screen is something most AI directors wouldn't even dare to dream of. Since the release of 'Niu Lai', countless OPCs are left feeling frustrated and introspective, because the benchmark has already been set. The production team consisted of just two people, with rumored production costs of only 1,680 yuan, excluding the computer, covering only minor expenses for drawing software and materials.

Of course, in terms of production costs, we can't ignore the fact that the mother-son duo spent five years frame-by-frame hand-crafting the film. Investors and the rent for the studio space likely wouldn't have been able to wait that long. Opening the long-neglected Maoyan Pro app, I found that 'Niu Lai' has already grossed 16.96 million yuan at the box office, with projected total earnings of 135 million yuan. What does this mean? The theoretical return on investment is 7.73 million times.

While it can't be compared to cinematic phenomena like 'Ne Zha: The Demon Boy Comes to the Sea' (15.446 billion yuan), in a specific year like 2026, within the animated film market, this is undoubtedly a standout achievement and the undeniable highlight of the year. On a global cinematic scale, this is also an impressive feat. The current recognized benchmark for ROI is 'Paranormal Activity' (2007), which used an ultra-low budget of $15,000 to secure approximately $194 million globally—an astonishing ROI of about 12,900 times. Compared to the same period, Nolan and Wen Muye also had films in theaters; their box office numbers may be winning, but they're likely envious of the overwhelming word-of-mouth support and relentless trending topics that 'Niu Lai' has generated.

Given that Seedance and its parent company ByteDance have made significant strides in AI video model capabilities and commercial revenue, I decided to pose my question to Doubao. The specific figures mentioned below are the values generated by Doubao's calculations. First, it's essential to clarify some basics: the most advanced AI video models currently cannot automatically produce a complete film. In AI video content, computing power is only a small part of the cost; human debugging, fixing visual continuity, and maintaining character consistency are the real expenses. Additionally, existing models typically have a waste rate of 3:1 to 5:1 per shot. The original 'Niu Lai' didn't use AI at all, with its simple visuals, straightforward shots, and minimal character movement, resulting in virtually zero post-production overhead. Using AI models would significantly inflate current costs.

If replicating a film of similar simple artistic style, narrative scale, and theatrical release standards (with proper certification) using an AI workflow, the budget can be divided into three tiers.

Where to begin

Tier 1: The extreme solo indie version (matching the primitive quality of 'Niu Lai', single-person production). Suitable for individual creators and experimental works that don't require polished, smooth visuals and allow slight shaking and occasional glitches. All work is done independently. Expense breakdown: AI video computing power/membership: 15,000-30,000 yuan, using domestic mid-tier text-to-video models (Keling, Jimeng, etc., subscription or pay-as-you-go). With about 860 shots per film, each shot generating an average of three versions for selection. Pre-and post-production labor (self-done, excluding salary, only tool costs): 4,000-8,000 yuan for AI painting memberships, voiceover and sound effect libraries, editing software, lip-sync tools, and upscaling tools; AI voiceovers and royalty-free background music. Mandatory theatrical delivery costs (essential for release): DCP theater master production and technical inspection: 6,000-8,000 yuan. Film registration is free; no marketing budget. Total budget: 25,000-40,000 yuan. Disadvantages: takes 3-6 months; frequent character drift and animation glitches require extensive manual patchwork; visual stability is likely inferior to the original 'Niu Lai'.

Tier 2: Small team professional version (recommended for stable quality, 2-3 person team). Goal: consistent character design, controllable motion coherence, reduced noticeable AI glitches, and basic viewing fluency, suitable for small-scale theater screenings. AI computing budget: 40,000-70,000 yuan, combining high-quality video models with increased key shot generation; extensive AI image-to-video and frame interpolation to repair broken shots. Outsourced labor costs (storyboard artists, AI prompt engineers, post-production editors/colorists, sound mixing): 80,000-150,000 yuan. This is the largest expense. Dedicated personnel are needed to lock character designs, fix AI-distorted limbs, ensure shot continuity, and handle lip-syncing. Copyrights, audio, commercial assets, and masters: 12,000-20,000 yuan. Total budget: 130,000-240,000 yuan. Timeline: 1.5-3 months; final quality is noticeably better than Tier 1 and can approach the completeness of an indie animated feature.

Tier 3: Industrial-standard high-end AI film (don't confuse this; it far exceeds 'Niu Lai''s positioning). To achieve commercial animation quality with consistent, stable output, combining large-scale AI generation with animators manually refining keyframes: 500,000-1.5 million yuan. According to Doubao, Tier 3 is no longer about replicating 'Niu Lai''s low-cost approach, so it won't be expanded upon. All three tiers have budgets far exceeding the 1,680 yuan spent on the original hand-crafted 'Niu Lai'.

Breaking down the real costs

I was curious, though, because industry leaders claim that AI blockbusters are about to hit theaters. So I pushed further, asking what it would actually cost to produce an industrial-standard AI animated film. Doubao analyzed two versions based on real 2026 domestic market conditions. Option A: The baseline for a mature domestic studio's theatrical AI animation, with a total budget of 500,000-1.2 million yuan, a 4-6 month production cycle, and an 8-12 person dedicated team. Option B: A quasi-commercial top-tier industrial AI workflow (approaching commercial theatrical animation quality), with a total budget of 1.5-3.5 million yuan, a 6-9 month cycle, and a team of 15-25 people, including directors, art directors, animation directors, technical artists, and sound supervisors.

Doubao also cautioned that once budgets exceed 4 million yuan, the industry widely believes cost-effectiveness collapses. Beyond that point, it's better to pivot to a "traditional hand-drawn + AI-assisted" model; pure AIGC native workflows struggle to match the expressive nuance of a multi-million-dollar traditional 2D animation. Additionally, there are three key industry truths: 1. Computing power is always just a small part; labor costs account for 60%-70%. 2. Don't fantasize about training proprietary large models to reduce long-term costs. 3. Copyright boundaries must be respected when drawing inspiration from 'Niu Lai'.

In reality, setting aside the cost and technical realities summarized by Doubao, every filmmaker or animator knows that even pouring millions more into a project wouldn't replicate the box office miracle of 'Niu Lai'. Not to mention, in terms of "artistic merit"... well, there are a thousand versions of 'Niu Lai' in a thousand people's eyes. Two people, one computer, and an extremely low cash outlay, yet they managed to secure a theatrical release. With its incredibly crude visuals, it struck a chord with emotions the industry least wants to discuss openly. At the core of this frenzy lies a widespread, gentle collective irony—a public disenchantment with and rebellion against mainstream narratives.

Since theatrical films became a market, in the public's simple value system, there's always been an effective market filter: bad films naturally go unnoticed. Yet people are constantly bombarded with content of varying quality, sloppy storytelling, and a lack of sincerity. On one hand, well-funded teams deliver mediocre results; on the other, crudely finished works manage to hit the big screen. Back in the real world, where absurdity has become the norm, people have accumulated a deep sense of powerlessness, giving rise to nihilism. Day-to-day life already forces us to accept various shortcomings—how can something so absurd happen? A work so crude that it's almost insulting manages to secure a theatrical release? Since we're already forced to accept so many unreasonable things in our daily lives, what's one more bad film? Unable to change the broader environment, 'Niu Lai' arrives at the perfect moment. As a symbol of absurdity, it reflects the mood of the era and serves as a release valve for collective pressure. This frenzy is a gentle, entertaining form of passive resistance. People cling to a safe symbol—one without major capital backing, one that won't offend the giants, and one confined to the realm of "movie entertainment"—using mockery and sarcasm to vent the exhaustion and disappointment buried deep inside.

From this perspective, the mother-son duo behind 'Niu Lai', who "silently bore it all," seem tragic. But they've already profited from the box office revenue share, so they're not exactly deserving of sympathy from ordinary folks like us. The world is too chaotic; AI has already dramatically lowered the barriers and costs of filmmaking. In the future, we may see more low-cost works like 'Niu Lai'. When that happens, will scenes like this become more common?