Research on the Innovation of Blended Course Construction Models Based on AIGC

Abstract

Generative artificial intelligence is transforming the way educational content is organized, and traditional blended courses find it difficult to adapt to its dynamism in terms of preset modules, linear interactions, and static evaluations. This study focuses on the innovation of AIGC-based blended course construction models and builds a framework from three levels: identifying the bottlenecks of traditional architectures, redefining course elements, and designing a three-layer human-machine collaborative structure; establishing a dynamic construction method for knowledge units driven by multimodal semantics, and achieving adaptive path orchestration with real-time feedback and generative models; promoting the shift of evaluation from outcome measurement to process tracking, integrating multiple data with AIGC, and constructing streaming analysis and double-loop closed-loop adjustment strategies. This model provides theoretical and technical pathways for the adaptive reconstruction of blended courses.