This study investigates the impact of autonomy, competence, relatedness, and excitement on user satisfaction with ChatGPT through the lens of Self-Determination Theory (SDT). Utilizing quantitative methods and a structured questionnaire, data from 332 university students in Taiwan with prior experience using ChatGPT were analyzed via Partial Least Squares Structural Equation Modeling (PLS-SEM). Results confirmed that usage frequency positively influenced autonomy, competence, relatedness, and excitement. Moreover, these psychological and emotional factors were found to significantly enhance user satisfaction. Among the variables, excitement demonstrated the strongest positive influence on satisfaction, underscoring the importance of emotional engagement in AI interaction contexts. Autonomy and relatedness also showed substantial positive effects, highlighting users’ preference for control and emotional connection in their interactions with AI tools. Competence, while significant, exhibited a comparatively weaker impact. These findings reinforce SDT’s relevance in examining AI-driven learning environments and offer practical implications for developers, emphasizing the need for user-centered designs that enhance emotional Excitement and psychological need fulfillment. Future research should explore deeper psychological processes, longitudinal changes in user engagement, and comparative studies across different AI platforms to extend the understanding and applicability of SDT in AI interactions.