The rapid advancement of Generative AI technology has positioned code generation services as key tools for enhancing productivity and creativity for both enterprises and individuals. This study empirically analyzes the factors influencing the continuous usage intention of code generation services based on Generative AI platforms. Drawing on the Information System Success Model and the Post-Acceptance Model, this research examines the pathways through which interactivity, reliability, ease of use, and code accuracy lead to satisfaction and continuous usage intention via expectation confirmation and perceived usefulness. Additionally, the impact of AI literacy, a personal user characteristic, on the acceptance process was identified. The analysis revealed that reliability and code accuracy significantly affected expectation confirmation, whereas perceived ease of use did not. This suggests that in code generation tasks, users prioritize functional correctness over interface convenience. This study provides strategic insights for quality management and fostering sustained user adoption of code generation services in the rapidly evolving AI landscape.
Kim et al. (Tue,) studied this question.