人工智能学习中的出版者权益保护路径探究
罗明东 1周安平2
作者信息
- 1. 西南大学法学院;西南大学出版与传播科学研究中心
- 2. 西南大学新闻传媒学院;西南大学出版与传播科学研究中心
- 折叠
摘要
出版业是致力于提供知识服务的行业,拥有丰富的出版物数据资源,是人工智能学习重要的数据来源.生成式人工智能作为一种颠覆性的技术,给出版业带来了巨大的挑战.人工智能学习中对出版物数据的复制、存储和预处理,存在侵犯出版者就出版物数据享有的版权等数据权益的风险.人工智能技术的发展和应用是大势所趋,需要为其学习数据的获取提供便利,但合理使用的路径将过度牺牲出版者的利益.因此,宜充分尊重出版者的数据权益,并探索建立集体管理等出版物海量许可模式.出版者也可发挥资源优势,采取技术措施进行自主维权.
Abstract
The publishing industry is the main force of content production and knowledge services and has the richest publication data resources which constitute the most important data source for artificial intelligence learning.Generative artificial intelligence,as a disruptive technology,has brought enormous challenges to the publishing industry.The replication,storage,and preprocessing of publication data in artificial intelligence learning pose a serious risk of infringing on the copyright and other data rights regarding publication data enjoyed by publishers.The development and application of artificial intelligence technology is an inevitable trend,and it is necessary to provide convenience for the acquisition of learning data.However,the path of fair use will excessively sacrifice the interests of publishers.Therefore,it is necessary to fully respect the data rights and interests of publishers,and explore the establishment of a massive licensing model for publications such as collective management.Publishers can also leverage their resource advantages and take technical measures to independently protect their rights.
关键词
人工智能/机器学习/文本和数据挖掘/数据库/出版者Key words
artificial intelligence/machine learning/text and data mining/data base/publisher引用本文复制引用
基金项目
2019年教育部人文社会科学研究青年基金西部和边疆地区项目(19XJC880005)
2023年重庆市文旅社科规划项目(2023WL09)
出版年
2024
NETL
NSTL
万方数据