본문 바로가기
  • Home

Generative AI based Customized Contract Clause Recommendation System for Game Content License Agreement

  • Journal of Software Assessment and Valuation
  • Abbr : JSAV
  • 2024, 20(4), pp.173-182
  • Publisher : Korea Software Assessment and Valuation Society
  • Research Area : Engineering > Computer Science
  • Received : November 17, 2024
  • Accepted : December 20, 2024
  • Published : December 31, 2024

Hyun-Soo Kim 1 Chang-Jun Choi 2 YongJoon Joe 3 Shin DongMyung 4

1엘에스웨어(주)
2엘에스웨어㈜
3엘에스웨어
4엘에스웨어 (주)

Accredited

ABSTRACT

This paper proposes an AI-based system for generating license agreement clauses customized to the characteristics of different game genres. Game content possesses unique traits depending on its genre, and these differences significantly influence contract terms, potentially increasing the probability of legal disputes. In this study, we fine-tuned a generative AI model, GPT-4o, to create personalized contract clauses optimized for specific game genres and contractual purposes. To achieve this, we analyzed publicly available standard contracts and addressed the lack of training data by expanding the dataset using TextGAN. Experimental results showed that the fine-tuned model, optimized through hyperparameter adjustments, achieved a decrease in Training Loss to 0.4635 and demonstrated improved performance in generating clauses suitable for game content license agreements compared to the base GPT-4o model. This system is expected to enhance the efficiency of contract drafting and reduce the potential for legal disputes.

Citation status

* References for papers published after 2023 are currently being built.