首页|人工智能生成物的法律性质及著作权侵权风险化解

人工智能生成物的法律性质及著作权侵权风险化解

扫码查看
[目的]对人工智能生成物的法律属性进行准确定位,寻找人工智能生成物的潜在风险化解路径.[方法]从法律层面分析人工智能生成物的产生来源及过程,揭示人工智能生成物的本质.[结果]从人工智能"数据训练"创作方式和机器学习阶段分析来看,人工智能不具备民事主体资格,但其作品具有可版权性的可能性,应被赋予一定权利.[结论]赋予人工智能享有署名权的权利,以"使人工智能生成物成型确定下来"为分界线,在自然人范围内确定权属和责任承担者进行权利安排.运用实质性相似规则和市场替代标准化解人工智能生成物复制权和改编权侵权风险,同时适用过错原则,相应配置人工智能设计者、使用者"防抄袭设计义务"注意义务.
The Legal Nature of Artificial Intelligence Products and the Resolution of Copyright Infringement Risk
[Purposes]This paper aims to accurately locate the legal attributes of artificial intelligence products and find the potential risk resolution path of artificial intelligence products.[Methods]The na-ture of artificial intelligence products is revealed through legal analysis of the source and process of artifi-cial intelligence products.[Findings]From the analysis of the creation method of artificial intelligence"data training"and the machine learning stage,artificial intelligence does not have the qualification of civil subject,but the works have the possibility of copyright,which should be granted certain rights.[Conclusions]This paper gives artificial intelligence the right to enjoy the right of signature,and with"the formation of artificial intelligence products to be determined"as the dividing line,this paper deter-mines the ownership and responsibility bearers within the scope of natural persons to arrange the rights.The substantive similarity rules and market substitution standards are used to resolve the risk of infringe-ment of reproduction rights and adaptation rights of artificial intelligence products.At the same time,the principle of fault is applied,and the duty of care of"anti-plagiarism design obligations"of artificial intel-ligence designers and users is configured accordingly.

artificial intelligence productsoriginalitysubstantial similaritypresumption of fault

东村仁、臧可

展开 >

河南财经政法大学知识产权学院,河南 郑州 450046

西北师范大学法学院,甘肃 兰州 730070

人工智能生成物 独创性 实质性相似 过错推定

2024

河南科技
河南省科学技术信息研究院

河南科技

影响因子:0.615
ISSN:1003-5168
年,卷(期):2024.51(6)
  • 23