China’s anime powerhouse Bilibili has made a significant move in the AI space with the launch of AniSora, a new open-source model built specifically for anime video generation. Drawing from a massive dataset of over 10 million clips sourced from 1 million animated videos, AniSora allows s to create fully animated scenes with just a single click. From manga adaptations to VTuber content, Bilibili promises for a wide range of anime styles—an ambitious offering that taps into both the creativity of fans and the demands of modern animation studios.

This announcement, from Anime Corner, positions Bilibili among the growing number of entertainment companies exploring AI to streamline the animation pipeline. As one of China’s most influential anime producers, with credits including Fire Force, Link Click, and The Heike Story, the company is no stranger to the challenges of time-intensive production. With AniSora, Bilibili hopes to offer creators new tools to speed up workflows and lower the barrier to entry, especially for indie teams and solo artists.

A Closer Look at AniSora’s Capabilities

Custom Control Meets Automation in AniSora

One Piece screenshots from the anime cut together showing Luffy Shocked while an injured Zoro and Giants with Shanks are in the background.

AniSora isn’t just a text-to-video generator, it is a versatile platform that s multiple creative workflows. s can generate videos from written prompts, static images, or even a sequence of frames to ensure smooth motion and continuity. One standout feature allows s to specify areas of motion within a scene, preserving still backgrounds while animating selected elements, like a character's expression or a flowing cape. This level of control mirrors the detail-conscious nature of traditional animation, but with far less manual labor.

Related
"A Violation of the Law": After ChatGPT's AI Attack on Studio Ghibli, Lawmakers Are Looking to Take Legal Action

Studio Ghibli has been the focus of discussions after AI started imitating its style, now Japanese lawmakers are questioning AI's legality.

19

Another key capability is multi-task learning, where AniSora can focus on specific animation tasks like lip sync in tandem with general motion. This is especially useful for dialogue-heavy scenes or VTuber content, where consistency in facial animation is critical. By offering reference-based video generation and localized motion control, AniSora sets itself apart from more generic AI models that often struggle with coherence across frames or maintaining a specific visual style.

Industry Implications and Ethical Tensions With This New AI

AI’s Growing Footprint in the Anime Industry

The launch of AniSora arrives during a time of heated debate over AI’s role in creative industries. On one hand, studios and independent creators alike could benefit from shorter production times and lower costs. Bilibili argues that its model addresses the animation industry’s demand for faster and more efficient tools, pointing out that manual processes like inbetweening and storyboarding have long been production bottlenecks. AniSora aims to automate parts of this workflow without entirely replacing human creativity.

However, the model’s reliance on unlicensed training data has sparked criticism. Since AniSora was trained on videos it did not create, some artists and studios worry that its outputs may directly compete with the works it learned from. Similar concerns have surfaced across the global industry, as AI threatens to blur the line between homage and exploitation in anime. While Bilibili s Netflix and KADOKAWA in pushing forward with AI R&D, others like Crunchyroll have taken a firm stance against such tools in anime production.