نوع مقاله : مقاله علمی - پژوهشی
نویسندگان
1 استادیار گروه حقوق تجارت بینالملل و حقوق مالکیت فکری و فضای مجازی، دانشکده حقوق، دانشگاه شهید بهشتی، تهران، ایران
2 دانشجوی دوره دکتری حقوق تجارت و سرمایهگذاری بینالملل، دانشکده حقوق، دانشگاه شهید بهشتی، تهران، ایران.
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
This paper examines the emerging copyrigh challenges arising from the rapid development of generative artificial intelligence, particularly large language models such as ChatGPT. The purpose of the study is to clarify the legal implications of AI advancements for the copyright system and to propose solutions for striking a balance between technological innovation and the protection of authors’ rights. The primary focus is on the legal dimensions of using copyrighted data in the training of these models. Through an analysis of recent lawsuits filed by copyright holders against AI system developers, the article explores how the doctrine of “fair use” has been interpreted and applied within the legal systems of the United States and the European Union. The findings reveal that while U.S. copyright law takes the rights of authors seriously, it nevertheless allows for fair use exceptions under certain circumstances, whereas EU regulations are more stringent and less inclined to accept such exceptions. Furthermore, given the increasing reliance on AI and the use of copyrighted data in Iran, the article assesses the current legal framework of Iran’s intellectual property system and highlights the legal gaps related to the use of copyrighted content in AI training. In Iranian law, not only is the concept of “fair use” undefined, but the existing frameworks also fail to adequately address the legal challenges posed by new technologies. A comprehensive revision is therefore necessary in light of these developments and in view of the country’s path toward advancing AI-related technologies.
کلیدواژهها [English]
)