🎙 Listen to a summary of this story
China’s campaign to turn artificial intelligence into an engine of economic growth is moving into film and online entertainment, as local governments offer subsidies, inexpensive workspaces and computing support to creators experimenting with AI-generated video.
The effort is rapidly lowering production costs and creating a new network of studios, technology companies and distribution platforms. It is also raising familiar concerns about overproduction, copyright infringement and the displacement of actors, animators and other creative workers.
Municipal authorities in Beijing, Shanghai and Shenzhen have introduced measures intended to attract AI filmmakers and related startups. The programmes include subsidised computing capacity, cloud-based models, rent reductions and assistance with overseas distribution. Hainan and other regions are offering incentives of their own.
Major entertainment companies are joining the expansion. Streaming platform iQIYI has offered subsidies to selected AI-content creators, while traditional studios are using generative systems to reduce the cost and time required to create characters, scenery and visual effects.
The shift reflects China’s broader strategy for artificial intelligence: encourage rapid experimentation, build regional industrial clusters and use public support to move promising technologies into commercial production. Similar policies helped China develop globally competitive electric-vehicle, battery and solar-panel industries.
In film, however, the product is not a manufactured component but a cultural work whose success depends on originality and audience interest. That makes the industry’s expansion more difficult to measure — and harder to direct through industrial policy.
Production costs fall as output surges
AI-video tools can generate animated sequences, synthetic performers and elaborate backgrounds without the crews, locations and equipment required for conventional filming. Chinese producers say scenes costing tens of thousands of yuan to shoot traditionally can sometimes be generated for a fraction of that amount.
According to figures cited by Chinese state broadcaster CCTV, the cost of producing AI-generated short dramas fell during the first half of 2026 from about 5,000 yuan per minute to several hundred yuan. Subsidised access to computing power can push costs lower still.
The declining price has produced an extraordinary increase in output. DataEye, a Chinese research company, counted 221,900 new AI programmes launched on Douyin during the first six months of 2026. Only 1,055 surpassed 100 million views, a widely used measure of commercial success.
Those figures suggest that the technology has made production easier without guaranteeing that audiences will watch what it creates. The same incentives that encourage experimentation may also generate a flood of similar programmes competing for limited attention.
China is nevertheless moving AI-generated work into mainstream cinema. The National Film Administration granted a public-screening licence to “Sanxingdui: Future Memories,” a 90-minute science-fiction production described as the first AI film from a major Chinese studio approved for theatrical release. Producer Bona Film Group has said the film is due for release this year.
The approval is an important test. A theatrical film must sustain characters and a narrative for far longer than the brief clips and micro-dramas that dominate social-media platforms. Its reception could show whether AI-generated video is developing into a durable creative form or remaining primarily a way to produce inexpensive online content.
Copyright rules lag behind the technology
The expansion is moving faster than China’s copyright framework. National rules require AI-generated text, images, audio and video to carry visible or embedded labels. Those measures are intended to help users distinguish synthetic material from conventionally produced content.
Labelling does not resolve who owns an AI-generated character, whether copyrighted footage may be used for training or when a synthetic performance unlawfully reproduces an actor’s face or voice. Chinese performers have expressed concern about unauthorised digital replicas and the potential loss of work for actors and voice artists.
Viewers have also complained about plagiarism, repetitive plots and programmes assembled from recognisable elements of existing works. Clearer copyright rules could determine whether the industry develops around licensed creative material or remains exposed to disputes over the origins of its images and characters.
The stakes extend beyond entertainment. China is trying to establish an early advantage in a technology that could reshape advertising, games, education and social media as well as cinema. Its enormous online audience gives developers a ready market in which to test formats and collect feedback.
But the emerging industry also illustrates the risks of China’s technology-driven growth model. Subsidies can build capacity quickly, yet they cannot ensure commercial demand or creative quality. If cities compete to establish nearly identical AI-video clusters, companies may depend more on public incentives than paying audiences.
For filmmakers, the attraction remains powerful: scenes once limited by budgets can now be generated on a laptop connected to cloud computing. For policymakers, the challenge will be turning that technical possibility into a sustainable industry without allowing cheap production to overwhelm originality, employment protections and intellectual-property rights.
