As a kind of "New quality productivity", text-to-video generative artificial intelligence has a broad application prospect in the field of book publishing. Compared with CHATGPT, text-to-video generative artificial intelligence Sora enables the construction of library visual digital resources, virtual space construction, intelligent service model innovation and user personalized service innovation. But at the same time, it also makes the library face the challenges of infringement of personal rights, intellectual property rights, data information, algorithm security and false information. In view of this, in order to eliminate these legal and technological risks, we should strengthen the effective management of the risks of the application of artificial intelligence Sora in the field of books, and clarify the ownership of the copyright of artificial intelligence products, and on this basis to further strengthen the algorithm supervision and enhance library's self-compliance ability so as to make text-to-video generative artificial intelligence technology better serve the sustainable development of the book industry, and jointly create a healthy and orderly human-computer coexistence environment.