| 英文摘要 |
In recent years, the application of generative artificial intelligence—characterized by its independence from human involvement, its ability to produce outputs identical to human creations, and its potential economic value—has sparked debates over whether such outputs should qualify for copyright protection. Divergent views on the requirements for copyright protection are reflected in both judicial opinions and academic perspectives. The overinclusive application of the originality requirement has, in turn, resulted in relatively little discussion of the requirement of“works of authorship.” The requirement of“works of authorship”consists of both the“author”and the“act of authorship,”and the categories of copyright subject matter. From the perspective of the author, some argue that the“human authorship requirement”is the most critical requirement for analysis. However,“authorship”cannot exist without an“act of authorship.” The requirement of“works of authorship”has traditionally relied on the theories of“conception and execution,”“authorship-by-adoption,”and“selection and arrangement,”all of which have given rise to ambiguities and shortcomings in application. The standards of“detailed conception”and“controlled execution”by the author cannot be universally applied to the diverse creative methods of today. Similarly, reliance on“authorship by adoption”and“selection and arrangement”renders it difficult to establish a causal connection between the author and the resulting work. This paper argues that the introduction of“authorship causation”—encompassing the elements of“intent to create,”“contribution in the act of authorship,”and“control over the act of authorship”—better accords with the foundations of intellectual property law, ensures applicability across diverse forms of authorship, and filters out circumstances that should not receive copyright protection. |