人工智能生成内容研究综述:应用、风险与治理
A Review of AI-generated Content Research:Applications,Risks,and Governance
李旭光 1胡奕 1王曼 1陆颖颖 1冯瑄2
作者信息
- 1. 山东理工大学信息管理学院 淄博 255000
- 2. 南开大学商学院信息资源管理系 天津 300071
- 折叠
摘要
[目的/意义]对人工智能生成内容(artificial intelligence generated content,AIGC)应用、风险及治理的研究,有助于把握人工智能未来的发展方向,为新兴主题研究提供参考与借鉴.[方法/过程]以中国知网、万方数据知识服务平台和Web of Science核心合集、ProQuest、ScienceDirect数据库为数据源,收集人工智能生成内容相关文献,分析AIGC在不同领域的应用现状及其风险,并提出AIGC的治理目标.[结果/结论]AIGC主要在医疗、新闻、学术、教育与艺术等领域取得一定成果,同时面临具有关联关系的来自技术、个人和社会3个层面的风险.本文通过构建"以人为本"的AIGC服务体系模型,从内部循环共生层面提出技术控制和个人提升的具体治理措施,从外部控制引导层面提出政策法规和社会引导具体措施,以指导生成式人工智能的设计、应用与发展.
Abstract
[Purpose/Significancel Research on the applications,risks,and governance of Artificial Intelligence Generated Content(AIGC)contributes to understanding the future development direction of artificial intelligence.It provides references for relevant emerging thematic studies.[Method/Process]Using CNKI(China National Knowl-edge Infrastructure),Wanfang,Web of Science,ProQuest,and ScienceDirect as data sources,this study collected literature related to AIGC,analyzed the current applications and associated risks of AIGC across various fields,and proposed governance objective for AIGC.[Result/Conclusion]AIGC has achieved certain accomplishments in fields such as healthcare,journalism,academia,education,and the arts,while concurrently facing interrelated risks from the technical,individual,and societal levels.This paper constructs a"Human-centered"AIGC service system model and proposes specific governance measures for technical control and personal enhancement at the internal cyclical symbio-sis level,as well as specific measures for policy formulation and social guidance at the external control guidance level,to guide the design,application,and development of generative artificial intelligence.
关键词
人工智能生成内容/生成式人工智能/以人为中心的人工智能/人机协同/治理目标Key words
artificial intelligence generated content/generative artificial intelligence/human-centered artifi-cial intelligence/human-machine collaboration/governance objective引用本文复制引用
基金项目
泰山学者工程专项()
国家自然科学基金面上项目(72272138)
出版年
2024
NETL
NSTL
万方数据