نوع مقاله : مقاله پژوهشی
نویسندگان
1 گروه حقوق خصوصی و اسلامی، دانشکده حقوق و علوم سیاسی، دانشگاه شیراز، شیراز، ایران
2 گروه حقوق خصوصی و اسلامی، دانشکده حقوق و علوم سیاسی، دانشگاه شیراز، شیراز، ایران.
3 دانشجوی دکتری حقوق خصوصی، دانشکده حقوق و علوم سیاسی، دانشگاه شیراز، شیراز، ایران.
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
The rapid development of generative artificial intelligence technologies has posed novel challenges to traditional intellectual property law frameworks. The use of copyrighted works and materials protected by related rights in training generative models demonstrates that existing legal regimes, particularly concerning related rights, are insufficient to address practical complexities. This study employs a comparative approach to examine the laws of the United States, the European Union, the United Kingdom, and Iran, analyzing how legal gaps, lack of clear protection for performers and phonogram producers, and the absence of effective mechanisms to identify infringements in AI training processes create legal uncertainty in the use of training data. The findings indicate that harmonizing the protection of training data and AI-generated works in Iran requires legislative reform and the establishment of clear legal frameworks, while policy measures based on international experiences can help bridge these gaps.
کلیدواژهها [English]