Procedural Content Generation (PCG) refers to creating content algorithmically as opposed to manually. PCG research is a sub-area of AI research, and existed long before deep neural networks popularized the term “Generative AI”. The goal of the PCG AI research program is to explore and use of AI techniques in support of the generating of content required for games, serious games and educational applications.
Published work:
Matthew McConnell, Richard Zhao. From Frustration to Fun: An Adaptive Problem-Solving Puzzle Game Powered by Genetic Algorithm. Proceedings of the Twenty-First AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE-25), Edmonton, Canada, November, 2025, p277-286. [Link] [Video Demo]
Mahdi Farrokhimaleki, Parsa Rahmati, Richard Zhao. From Unstable to Playable: Stabilizing Angry Birds Levels via Object Segmentation. Proceedings of the Twenty-First AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE-25), Edmonton, Canada, November, 2025, p227-236. [Link]
Mahdi Farrokhimaleki, Richard Zhao. Procedural Content Generation in Games: A Survey with Insights on Emerging LLM Integration. Proceedings of the Twentieth AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE-24), Lexington, USA, November, 2024. [Link]
Extended Abstract:
Matthew McConnell, Richard Zhao. A Demonstration of Pathfinding-Based Puzzle Generation with Adaptive Difficulty. Proceedings of the Twenty-First AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE-25), Edmonton, Canada, November, 2025, p396-398. [Link] [Video Demo]
