China built the AI content factory that Silicon Valley imagined but never shipped

In January 2026, a new AI-generated micro-drama went live on a Chinese streaming platform every 90 seconds. By March, approximately 50,000 AI-native titles had been added to Douyin in a single month. The production cost was roughly one-tenth of a live-action shoot. The usable rate of AI-generated footage had climbed above 90 per cent. And the Chinese government was subsidising the whole thing. China’s micro-drama industry, a format of serialised stories told in episodes of one to three minutes and viewed vertically on a phone, is projected to exceed 120 billion yuan this year, surpassing the country’s entire theatrical box office for the first time. It is a $16.5 billion industry serving 660 million users, and it has become, without anyone in Silicon Valley quite noticing, the world’s first mass commercial application of AI-generated video. OpenAI shut down Sora in March after six months of commercial irrelevance. In the same month, China’s AI video industry produced more new titles than Netflix has released in its entire history.

The factory

The micro-drama production model that emerged in China around 2023 was already designed for speed. Studios operated in clusters of compact sets, known informally as “shudian,” or vertical studios, a play on the name of Hengdian, China’s largest traditional film production complex. Within a single facility, sets for hospitals, mansions, subway platforms, and banquet halls sat side by side, allowing crews to shoot scenes in rapid succession with minimal setup time. A typical live-action micro-drama could be produced in six to eight weeks for a few hundred thousand yuan. The economics were brutal but functional: low cost, high volume, algorithmic distribution through ByteDance’s Red Fruit platform, Tencent’s WeChat Video Accounts, and Kuaishou’s Xi Fan brand, with revenue generated through in-app purchases, advertising, and subscription models.

AI has compressed this cycle further. Companies have begun allocating approximately 30 per cent of their production budgets to AI-driven workflows, cutting total production time from three months to one and reducing costs to roughly one-fifth of traditional shoots. The shift is visible in the numbers. In January 2026’s top 100 micro-drama chart, AI-generated titles accounted for 38 per cent, up from 7 per cent a year earlier. DataEye, the industry tracking firm, reported that more than 10,000 AI-generated animated micro-dramas were going live each month by the start of 2026. AI-generated comic-style micro-dramas alone represented an estimated 16.8 billion yuan in market value in 2025. The tools driving this expansion, ByteDance’s Seedance 2.0, Kuaishou’s Kling 3.0, and Shengshu Technology’s Vidu, have pushed the quality of generated footage to the point where, for certain genres, the difference between AI-produced and live-action content is no longer immediately obvious to viewers scrolling through their feeds.

The state

What distinguishes China’s AI micro-drama boom from other AI-generated content experiments is the role of the state. Local governments across China have established production hubs in second and third-tier cities, offering subsidies, infrastructure, and talent incentives to attract micro-drama companies. Chongqing built the Liangjiang Film and Television Animation Cultural and Creative Park, a facility designed specifically for vertical-format content that now hosts more than 300 production crews annually. The city has been dubbed “Vertical Hengdian” by industry participants. Linping, in Zhejiang province, allocated over 100 million yuan to support content creators and constructed China’s first dedicated micro-drama film base. The maximum guaranteed state incentive can reach two million yuan per individual drama, with additional support structures covering funding, content development, data access, and traffic allocation.

The state’s interest is not purely cultural. In January 2026, CCTV, China’s state broadcaster, produced its own AI-generated comic drama, a signal that the technology had been endorsed at the highest level of official media. The National Radio and Television Administration has simultaneously tightened its regulatory framework, establishing a tiered content review system based on production budgets. Productions exceeding one million yuan require provincial-level approval. Mid-tier productions face a separate review track. Smaller productions are overseen by the hosting platforms themselves. The NRTA has removed more than 25,000 episodes for content violations, and it has issued specific guidance on animated micro-dramas, requiring pre-broadcast review and platform-based content checks. The regulatory architecture serves a dual purpose: it encourages the industry’s growth as an economic engine and a tool for cultural export, while ensuring that the content produced remains within the boundaries of what the state considers acceptable. It is industrial policy applied to entertainment, and it is working.

The quality problem

The micro-drama industry’s relationship with AI mirrors a pattern that has emerged across every sector where generative AI has lowered the barriers to production. When the cost of creating something falls by 90 per cent, the volume of what gets created increases by an order of magnitude, and most of it is mediocre. The same dynamic has played out in software, where AI coding tools have produced an explosion of applications with systemic quality problems. In China’s micro-drama market, the consequence is intense competition at the lower end and significant content homogeneity: thousands of titles with similar plots, similar visual styles, and similar emotional beats, all generated from the same underlying models and optimised for the same algorithmic distribution systems.

The industry’s response has been to bifurcate. Hongguo, one of the leading micro-drama production companies, announced plans to raise its overall content budget by more than 40 per cent in 2026 while maintaining investment in live-action production alongside its AI-generated output. The strategy reflects a recognition that AI-generated content serves a different market segment from live-action micro-dramas, and that the highest-revenue titles still tend to be those with human actors, original scripts, and production values that distinguish them from the algorithmically generated flood. For genres such as mythological fantasy and historical stories, where visual effects have traditionally been expensive, AI offers clear advantages. For genres that depend on emotional nuance, physical performance, and the particular quality of a human face, the technology remains a complement rather than a replacement.

The export

China’s micro-drama industry is no longer a domestic phenomenon. In the first eight months of 2025, overseas micro-drama revenue reached $1.525 billion, a 195 per cent year-on-year increase. ReelShort, DramaBox, GoodShort, and My Drama have established significant user bases in the United States and Southeast Asia, with ReelShort alone generating approximately $400 million in 2024 revenue, though it remains unprofitable due to heavy marketing investment. The global micro-drama market reached an estimated $11 billion in 2025 and is projected to hit $14 billion in 2026. Harvard Business Review published “Lessons from China’s Short-Drama Boom” in March 2026, describing the industry as a complete operating system: a production methodology, an algorithmic growth engine, and a modular monetisation stack honed in one of the world’s most competitive digital markets.

The comparison with Quibi, the American short-form streaming service that burned through $1.75 billion and shut down in 2020 after less than a year, is instructive. Quibi tried to create short-form video by slicing Hollywood-style content into smaller pieces for mobile viewing. China’s micro-drama industry did the opposite: it built an entirely new production model designed from the ground up for vertical screens, algorithmic distribution, and consumption in fragmented, informal bursts. The addition of AI has accelerated this model’s advantage. China has demonstrated repeatedly that it can match or exceed Western AI capabilities at a fraction of the cost, and the micro-drama industry is the most commercially visible example of that principle applied to content creation at industrial scale.

The test

What China’s micro-drama industry represents is not an entertainment trend but an industrial experiment. It is the first large-scale test of what happens when AI-generated content becomes a commodity: produced in volume, distributed algorithmically, consumed passively, regulated by the state, and exported globally. The economics are compelling. The quality is uneven. The cultural implications are largely unexamined. And the broader pattern of AI transforming creative production is not unique to video. Every industry that produces content, from software to journalism to advertising, is facing the same question: what happens when the cost of production falls to near zero and the volume of output becomes effectively infinite?

China’s answer, at least in micro-dramas, is that the state steps in to fund the infrastructure, regulate the output, and promote the export. The private sector provides the platforms and the AI models. The audience provides the attention. And the content itself, thousands of new titles every month, each one to three minutes long, each one optimised for a thumb scrolling through a feed, becomes the raw material of an entertainment economy that bears no resemblance to the one that Hollywood built and that Silicon Valley assumed it would disrupt. The disruption came from a different direction entirely, and it arrived on a vertical screen.