生成式人工智能的版权风险及其治理  

Copyright Risks and Governance of Generative Artificial Intelligence

作　　者：陶中怡 TAO Zhongyi(School of Law,Huaiyin Normal University,Huai'an,Jiangsu 223001)

机构地区：[1]淮阴师范学院法学院,江苏淮安223001

出　　处：《淮阴师范学院学报（哲学社会科学版）》2024年第6期573-580,647,共9页Journal of Huaiyin Teachers College(philolophy and social sciences Edition)

基　　金：2019年度江苏省高校哲学社会科学基金项目“大数据环境下文本与数据挖掘的版权问题研究”(2019SJA1623)。

摘　　要：生成式人工智能在一定程度上改变了传统的创作过程,从而对既有的著作权规则产生了冲击。生成式人工智能在数据获取和挖掘阶段是否适用合理使用著作权例外存在争议,在内容生成阶段存在可版权性判断规则的争议,在内容生成后的使用阶段存在侵权风险。为应对这些问题,有必要适当增加人工智能公司训练数据来源披露义务,在现有的立法框架下借助司法裁判认定数据获取和挖掘的法律性质,并鼓励人工智能企业与著作权集体管理组织形成合作,努力达成作者、著作权人、人工智能企业和用户等多方主体之间利益平衡。Generative Artificial Intelligence(GAI)has to some extent changed the traditional creative process,thereby impacting existing copyright rules.There is controversy over whether GAI applies fair use in the data acquisition and mining stages,whether the generated content is protected by copyright,and the risk of infringement in the use stage after content generation.To address these issues,it is necessary to appropriately increase the disclosure obligation of training data sources for artificial intelligence companies,use judicial rulings to determine the legal nature of data acquisition and mining within the existing legislative framework,and encourage artificial intelligence enterprises to cooperate with copyright collective management organizations,so as to achieve a balance of interests among multiple parties such as authors,copyright owners,artificial intelligence enterprises,and users.

关 键 词：生成式人工智能 数据挖掘 版权 风险治理 

分 类 号：D923[政治法律—民商法学] TP18[政治法律—法学]