Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

GPT for Games: An Updated Scoping Review (2020-2024)

Daijin Yang, Erica Kleinman, Casper Harteveld

TL;DR

This paper addresses how GPT and related large language models are transforming games by surveying 177 studies from 2020–2024, with a major 2024 surge. It identifies five core use cases—procedural content generation, mixed-initiative design and gameplay, autonomous gameplay, and game-user research—while documenting methodological trends, model choices, and interaction patterns. The authors highlight advances in prompt engineering, retrieval-augmented generation, and multi-agent architectures, yet note persistent challenges in coherence, long-term planning, and safety. They call for broader game-domain testing, open-weight models, and user-centered evaluations to better harness GPT's potential for richer, more adaptive gaming experiences. This work provides a foundation for researchers and developers aiming to integrate GPT into game design, testing, and player engagement in scalable, responsible ways.

Abstract

Due to GPT's impressive generative capabilities, its applications in games are expanding rapidly. To offer researchers a comprehensive understanding of the current applications and identify both emerging trends and unexplored areas, this paper introduces an updated scoping review of 177 articles, 122 of which were published in 2024, to explore GPT's potential for games. By coding and synthesizing the papers, we identify five prominent applications of GPT in current game research: procedural content generation, mixed-initiative game design, mixed-initiative gameplay, playing games, and game user research. Drawing on insights from these application areas and emerging research, we propose future studies should focus on expanding the technical boundaries of the GPT models and exploring the complex interaction dynamics between them and users. This review aims to illustrate the state of the art in innovative GPT applications in games, offering a foundation to enrich game development and enhance player experiences through cutting-edge AI innovations.

GPT for Games: An Updated Scoping Review (2020-2024)

TL;DR

This paper addresses how GPT and related large language models are transforming games by surveying 177 studies from 2020–2024, with a major 2024 surge. It identifies five core use cases—procedural content generation, mixed-initiative design and gameplay, autonomous gameplay, and game-user research—while documenting methodological trends, model choices, and interaction patterns. The authors highlight advances in prompt engineering, retrieval-augmented generation, and multi-agent architectures, yet note persistent challenges in coherence, long-term planning, and safety. They call for broader game-domain testing, open-weight models, and user-centered evaluations to better harness GPT's potential for richer, more adaptive gaming experiences. This work provides a foundation for researchers and developers aiming to integrate GPT into game design, testing, and player engagement in scalable, responsible ways.

Abstract

Due to GPT's impressive generative capabilities, its applications in games are expanding rapidly. To offer researchers a comprehensive understanding of the current applications and identify both emerging trends and unexplored areas, this paper introduces an updated scoping review of 177 articles, 122 of which were published in 2024, to explore GPT's potential for games. By coding and synthesizing the papers, we identify five prominent applications of GPT in current game research: procedural content generation, mixed-initiative game design, mixed-initiative gameplay, playing games, and game user research. Drawing on insights from these application areas and emerging research, we propose future studies should focus on expanding the technical boundaries of the GPT models and exploring the complex interaction dynamics between them and users. This review aims to illustrate the state of the art in innovative GPT applications in games, offering a foundation to enrich game development and enhance player experiences through cutting-edge AI innovations.

Paper Structure

This paper contains 16 sections, 3 figures.

Table of Contents

  1. Introduction
  2. Methodology
  3. Results
  4. General GPT Usage Trends of GPT for Games
  5. General Research Trends of GPT for Games
  6. Use Cases of GPT for Games
  7. GPT for Procedural Content Generation (PCG)
  8. GPT for Mixed-Initiative Game Design and Development
  9. GPT for Mixed-initiative Gameplay
  10. GPT as a Game Player
  11. GPT for Game User Research
  12. Discussion
  13. Expanding the Technical Boundaries of the Models
  14. Exploring the Complex Interaction Dynamics between Users, GPT, and its Generated Content
  15. Limitation
  16. ...and 1 more sections

Figures (3)

  • Figure 1: Flow Diagram for Systematic Search and Screening. The number before the plus sign shows previously identified articles, and the number after shows newly identified ones.
  • Figure 2: The General GPT Usage Trends of GPT for Games. 122 new studies were published in 2024, up from 39 in 2023, showing a significant growth trend compared to previous years. There was also rising interest in the latest GPT models like GPT-3.5 and GPT-4. One paper 142 did not mention the GPT version it used.
  • Figure 3: The General Trends of Each Category. In 2024, most papers focused on PCG and MIG, with growing interest in MIGDD, PG, and GUR. Three papers were counted in both PCG and MIG. One paper was counted in both PCG and GUR. In all, 56 papers focused on PCG, 39 on MIGDD, 52 on MIG, 20 on GPT playing games, and 14 on GUR.