AI生成物的著作权的法学研究
Legal Research on Copyright of AI-Generated Works

随着人工智能技术的快速发展，AI生成作品在文学、艺术、音乐等领域的应用日益广泛，但其著作权归属与保护问题引发学界与实务界的广泛争议。生成式AI主要通过使用者“投喂”一些数据，根据关键指令，利用深度学习技术来生成文本、图像、音频等内容。这些人工智能生成物与人类创作的作品极其相似，人们很容易产生混淆，从而在司法实践中对著作权归属的判断产生误导。本文以AI生成物的著作权研究为核心，基于著作权法理论，分析AI生成物的“独创性”认定标准，结合武汉“AI生成图侵权案”等司法实践系统探讨AI生成内容的权利主体认定。
With the rapid advancement of artificial intelligence (AI) technology, AI-generated works have been increasingly utilized in literature, art, music, and other fields. However, controversies persist in academia and legal practice regarding the ownership and protection of copyright for such works. Generative AI systems primarily rely on users “inputting” data and generating text, images, audio, and other content through deep learning technologies based on key instructions. These AI-generated outputs bear striking resemblance to human-authored works, leading to potential confusion and misleading implications in judicial determinations of copyright ownership. Focusing on the copyright issues of AI-generated content, this study analyzes the “originality” criteria for such outputs under copyright law theories, and systematically explores the identification of rights holders through judicial cases such as the “Wuhan AI-generated Image Infringement Case”.