Artificial Intelligence for Text Generation: An Intellectual Property Perspective

Purpose: This paper aims at proposing relevant improvement measures for the intellectual property regulation on AI-generated text. Methods: This paper adopts qualitative research methods, collects and analyzes opinions from various sources to illustrate the technical logic of AI-generated text and explore existing intellectual property legal regulations and dilemmas. Results: AI-generated content, especially AI-generated text, follows the process of data collection and processing, models creation and content generation. Regarding intellectual property regulations on the AI-generated text, international law reflects hysteresis while major states’ domestic law varies in terms of intellectual property attribution. Conclusion: Based on the research on current situations, this paper proposes corresponding improvement measures, namely self-censorship, self-improvement, international cooperation and strengthening legislation.