Abstract: Based on the qualitative research method and grounded theory, nine main categories have been extracted and summarized, including“ the improvement and optimization of intelligent content generation, the acceleration and upgrading of industrial production efficiency, the delivery and dissemination of personalized customized content, the formulation of integrated publishing channel strategies, the analysis and insight of commercial market data, the development and innovation of new business models, the transformation and reorganization of organizational structure and processes, the maintenance and creation of brand core values, and the consideration of legal and regulatory ethical risks”, as well as one core category of“ transformation and upgrading of traditional publishing enterprises”. Based on the analysis of the typical relationship structure between the main categories and the core category, as well as the logical relationship between categories, a theoretical model of generative artificial intelligence is constructed to empower traditional publishing enterprises to transform and upgrade effectively. The model reveals that how to better empower the productivity improvement of the backend and the creation of new value in the frontend with generative artificial intelligence models is the key to the transformation and upgrading of traditional publishing enterprises.