AI-Empowered Model Innovation and Adaptation in Innovation and Entrepreneurship Education: An Exploration Based on Multimodal Learning Scenarios

Abstract

Against the backdrop of the deep integration of "AI + Education" and the strategic upgrade of "Mass Entrepreneurship and Innovation", traditional innovation and entrepreneurship education faces challenges such as insufficient personalized training and monotonous practical scenarios. This study focuses on innovative pathways for AI-empowered education, constructing a theoretical framework based on multimodal learning scenarios and employing fuzzy-set qualitative comparative analysis (fsQCA) to uncover the key mechanisms driving educational model transformation under AI. The research finds that AI, through tools such as intelligent content generation, cross-modal interaction technology, and virtual practice simulation, enhances teachers' and students' AI literacy and technical adaptability while ensuring data security. This enables the construction of a trinity educational ecosystem—"data sensing, personalized adaptation, and resource sharing"—effectively addressing the disconnection between theory and practice in traditional education.