In recent years, generative artificial intelligence (AIGC) has experienced rapid development and widespread application in the publishing sector. As an emerging technological application and new quality productivity, AIGC is reshaping the publishing ecosystem, catalyzing profound transformations in content production and publishing workflows. How to deal with the subversive changes brought by generative artificial intelligence to the publishing field is an important issue that needs urgent academic attention and analysis. Existing studies have explored its application prospect, potential risks, ethical compliance, copyright protection, governance and supervision of AIGC in the publishing industry. Based on literature analysis, this paper focuses on the technical logic and implementation paths of generative artificial intelligence empowering publishing industry from the perspective of marketing theory. The application of generative artificial intelligence in publishing marketing enables precise market predictions and facilitates efficient decision-making in publishing marketing. Leveraging robust data processing and self-learning capabilities, it constructs precise user profiles by analyzing multidimensional data, including user behavior and purchasing preferences. Furthermore, generative artificial intelligence autonomously generates personalized textual content aligned with user preferences, thereby elevating reading experiences. Through model training based on user portraits, the algorithm can accurately identify user demands across various scenarios, thereby enabling more precise and personalized publishing products recommendations. Regarding specific practical paths, this paper proposes that generative artificial intelligence plays a significant role in the planning and text processing of publishing topics, personalized promotion and targeted distribution of publishing marketing materials, automated marketing copy generation, and creative design of book industry advertisements, as well as intelligent marketing customer service and emotional services. First, in topic planning, generative artificial intelligence builds an intelligent publishing topic planning system through multidimensional data fusion analysis, enhancing editorial efficiency. Second, in text content processing and optimization, AIGC's multi-modal depth fusion ability can realize the intuitive transformation of text content into visual images. Third, based on user portraits, generative artificial intelligence can predict user preferences and interests in advance and accurately recommend book products through precise identification and matching. Fourth, on the basis of generating the book content summary, publishing operators can also utilize AIGC to generate book covers and summaries, and further explore and refine the marketing selling points of the book, thus enhancing the marketing advantages of the book. Additionally, AIGC-powered intelligent customer services and emotional engagement enhance user emotional experience, and build deep trust between enterprises and users. As emerging technology is a "double-edged sword", how to avoid the potential risks that may be brought by generative artificial intelligence requires careful consideration. This paper advocates for balanced implementation with actively promoting AIGC applications in publishing and marketing, while mitigating potential risks in applications. Therefore, adherence to the principles of "ethics first" and "technology for good" is essential, embedding ethical principles such as fairness, justice, and honesty into model training to promote the transformation of publishing marketing toward a new paradigm of "human-machine collaboration".