With the development of the digital economy and artificial intelligence (AI), traditional business education faces pressure for transformation. The industry’s demand for interdisciplinary business talents with both digital intelligence literacy and practical capabilities is growing rapidly. However, traditional courses suffer from bottlenecks such as outdated knowledge updates, weak practicality, insufficient personalized training, and a single evaluation system, leading to a structural mismatch between talent supply and demand. Focusing on the deep integration of AI and business teaching, this study constructs an AI-enabled closed-loop teaching model of “learning status diagnosis—path planning—teaching implementation—intelligent evaluation—dynamic optimization” based on constructivism and cognitive load theory, and proposes three innovative paths: curriculum content reconstruction, human-machine collaborative teaching, and process-oriented intelligent evaluation. Practical verification shows that this model can effectively solve the pain points of traditional teaching, significantly improve teaching efficiency, help increase the approval rate of students’ national innovation and entrepreneurship projects by 50%, and raise the employment matching degree for AI-related positions by 35%. Meanwhile, this study sorts out the potential risks of technology implementation and puts forward targeted coping strategies, providing a replicable and promotable practical path for the digital transformation of business education in the new era.