arXiv:2403.02454v1 Announce Type: new
Abstract: This paper studies how large language models (LLMs) can act as effective, high-level creative collaborators and “muses” for game design. We model the design of this study after the exercises artists use by looking at amorphous ink splotches for creative inspiration. Our goal is to determine whether AI-assistance can improve, hinder, or provide an alternative quality to games when compared to the creative intents implemented by human designers. The capabilities of LLMs as game designers are stress tested by placing it at the forefront of the decision making process. Three prototype games are designed across 3 different genres: (1) a minimalist base game, (2) a game with features and game feel elements added by a human game designer, and (3) a game with features and feel elements directly implemented from prompted outputs of the LLM, ChatGPT. A user study was conducted and participants were asked to blindly evaluate the quality and their preference of these games. We discuss both the development process of communicating creative intent to an AI chatbot and the synthesized open feedback of the participants. We use this data to determine both the benefits and shortcomings of AI in a more design-centric role.
AI as a Creative Collaborator in Game Design
In this paper, the authors explore the potential of large language models (LLMs) to act as effective creative collaborators and sources of inspiration for game design. Drawing from the exercises artists use with ink splotches, the goal of the study is to investigate whether AI assistance can enhance, hinder, or provide an alternative creative quality to games compared to human designers.
The study focuses on stress testing the capabilities of LLMs as game designers by placing them at the forefront of the decision-making process. Three prototype games are designed across different genres, each with a different approach: a minimalist base game, a game with human-designed features and game feel elements, and a game with features directly implemented from prompted outputs of the LLM, ChatGPT.
A user study is conducted, where participants are asked to blindly evaluate the quality and their preference of these games. The paper not only discusses the development process of communicating creative intent to an AI chatbot but also presents the synthesized open feedback from the participants. This data is then analyzed to identify the benefits and shortcomings of AI when taking on a more design-centric role.
The Multi-Disciplinary Nature of the Concepts Explored
This study explores the intersection of AI, game design, and human creativity, making it highly multi-disciplinary. It combines the fields of artificial intelligence, human-computer interaction, and game development. By incorporating AI as a creative collaborator, the authors bridge the gap between technology and artistic expression, opening up new possibilities for game design.
On the AI side, the study leverages large language models to generate game features and elements. This showcases the potential of AI in understanding human-created content and generating novel ideas based on that input. By involving AI in the decision-making process, the study also raises questions about the role of AI in shaping the future of creative industries.
From a game design perspective, this research introduces an alternative approach to the traditional designer-centric process. It highlights the potential benefits of using AI as a source of inspiration and creativity. The study’s findings could lead to new methods of game design that blend the expertise of human designers with the innovation of AI.
Expert Analysis and Insights
This study presents an important step towards understanding the role of AI in creative processes. By involving LLMs in game design, the authors demonstrate how AI can contribute to, enhance, and even challenge human creativity. While AI can offer unique and unexpected ideas, it still falls short in fully reproducing the depth and richness of human-generated content.
The user study conducted in this research provides invaluable insights into the perceptions and preferences of players. It reveals that participants have different preferences when it comes to games created with human design intervention versus those designed solely based on AI-generated features. This suggests that a hybrid approach, combining AI-generated suggestions with human expertise, could be the most promising direction for future game design processes.
One limitation of this study is that it focuses on only three prototype games and their evaluation by a specific user group. Expanding the research to a larger set of games and a more diverse set of participants could provide further insights into the generalizability of the findings.
In conclusion, this research highlights the potential of AI as a creative collaborator in game design. It explores the benefits and limitations of involving AI in the decision-making process and provides valuable insights for the future of AI-assisted game development. By embracing the multi-disciplinary nature of this study, the authors pave the way for further exploration of AI’s role in creative industries.