Artificial intelligence is developing rapidly, and this technological advancement has led researchers to focus on how AI impacts students’ creative learning performance, especially in solving creative problems. In this context, this study combines the Technology Acceptance Model (TAM) and diary research methods to systematically explore university students’ use of ChatGPT to assist in solving creative problems, and to analyze the factors influencing its acceptance. The quantitative findings reveal that Perceived Ease of Use, Perceived Usefulness, Self-Efficacy, and Collaborative Learning have directly or indirectly significant impacts on students’ Behavioral Intention. The qualitative diary study, on the other hand, highlights the advantages students experienced during their interaction with ChatGPT, such as improved information integration efficiency, enhanced empathy, and increased confidence and sense of achievement. However, it also exposed limitations in the interaction process, such as inconsistent emotional expression, insufficient information accuracy, and challenges in prompt design. Future research should focus on improving AI’s emotional simulation capabilities, enhancing the accuracy of information, and simplifying prompt design to further enhance the effectiveness of AI in fostering creativity. The findings of this study provide both theoretical and practical support for the optimization and application of AI tools in educational practice, promoting further integration of AI technology and education.