摘要
人工智能数据训练过程中涉及作品的获取与利用,具有侵害著作权的风险,并且难以适用现行合理使用规则和法定许可制度。将人工智能数据训练纳入著作权合理使用范畴是促进人工智能产业发展和文学艺术繁荣发展的需要,符合利益平衡原则和市场失灵的法经济学理论,且有国际经验可供借鉴,国内司法判例已为立法提供方向。在《中华人民共和国著作权法实施条例》中增设人工智能数据训练合理使用条款是化解人工智能数据训练侵权风险的现实可行的路径。该条款中的使用主体应涵盖商业主体、科研机构和文化遗产机构;使用目的应限制为非表达性目的使用和一般性表达使用;使用方式应以不损害著作权人的合法权益为必要。
The acquisition and utilization of works involved in the training process of artificial intelligence data carry the risk of copyright infringement and are difficult to apply fair use rules and legal licensing systems.Incorporating the training of artificial intelligence data into the category of fair use of copyright is necessary for promoting the development of the artificial intelligence industry and the development of literature and art.This measure is in line with the principle of balance of interests and the legal and economic theory of market failure and has international experience to draw upon;moreover,domestic judicial precedents have provided direction for legislation.Adding a fair use clause for artificial intelligence data training in the"Regulations for the Implementation of the Copyright Law of the People's Republic of China"is a realistic and feasible way to defuse the infringement risk of artificial intelligence data training.The subjects using this clause should cover commercial entities,scientific research institutions and cultural heritage institutions.The purpose of use should be limited to non-expressive and general expressive uses,and the manner of use must not unreasonably prejudice the legitimate rights and interests of the copyright owner.
作者
朱文玉
陈丽敏
ZHU Wenyu;CHEN Limin(College of Humanities and Law,Northeast Forestry University,Harbin 150040,China)
出处
《太原师范学院学报(社会科学版)》
2026年第1期46-53,共8页
Journal of Taiyuan Normal University:Social Science Edition
基金
黑龙江省博士后出站科研启动基金项目“数据利用和保护法律问题研究”(415937)。
关键词
人工智能数据训练
著作权
合理使用
规则
artificial intelligence training data
copyright
fair use
rules
