Generative AI in 2D Animation: Applications, Challenges, and Future Prospects

Generative AI has been developing quickly in recent years and is able to influence different stages of 2D animation production. This paper reviews the current applications of generative AI in 2D animation from the perspective of computer graphics, its main challenges, and future prospects. The paper discusses how generative AI supports pre-production visual development, which includes concept art, character design, storyboard, and layout generation. It also reviews its applications in motion generation and frame synthesis, which include pose-guided animation, in-betweening, and animation sequence generation. The paper also analyzes its role in visual refinement and end-to-end generation, which includes colorization, style control, and integrated animation production systems. This paper identifies three challenges: limited controllability and interpretability, weak temporal and stylistic consistency, and limited applicability in real animation workflows; it outlines three future directions: more controllable generation frameworks, improved long-range consistency and style preservation, and better integration into artist-centered production pipelines. Overall, generative AI shows strong potential as a helpful tool in 2D animation production, but it still cannot fully replace human artists in the whole workflows.