A Study on a New Metric for Measuring the Security and Utility of Pseudonymized Data

The development of industries utilizing AI is growing so rapidly that it is difficult to predict the future.Generative AI produces results so quickly that it surpasses the labour time humans invest in creative activities.The areas related to management and service are sectors where content-generating AI can play an important role.Therefore, this study aims to analyze the impact of content-generating AI service quality on customer satisfaction and reuse intention.The factors determining service quality are accessibility, tangibility, responsiveness, and assurance.To achieve the research objective, a survey targeted adults in their 20s and older who use content-generating AI.The structural model for basic statistical analysis and hypothesis verification of the survey was implemented using Smart PLS 4.0.The results of the study are summarized as follows.First, accessibility, tangibility, responsiveness, and assurance had a significant effect on customer satisfaction.Second, customer satisfaction had a significant effect on reuse intention.In this study, users of content-generating AI evaluated reuse intention in accessibility, responsiveness, tangibility, and assurance among the service quality factors.These results suggest that content-generating AI businesses should enable users to easily access information and services and provide information promptly when needed.This emphasizes the importance of accessibility and responsiveness in service quality to optimize user experience and enhance customer satisfaction.However, this study has limitations in the possibility of age bias, as the proportion of respondents in their teens and those in their 50s or older was low.Follow-up research on various user groups is needed to supplement this limitation.This study has limitations due to the low proportion of respondents in their teens and those aged 50 or older, which may result in age bias.To address this limitation, follow-up research involving diverse user groups is necessary.