首页|生成式人工智能大模型全球治理的理论证成与初步构想

生成式人工智能大模型全球治理的理论证成与初步构想

扫码查看
生成式人工智能作为人工智能领域重要技术分支,不仅起到驱动数字经济变革和推进产业智能化升级的正面效应,也引发若干新型社会问题和国家安全风险,生成式人工智能大模型全球治理的提出正当其时.欧盟作为数字时代中美之外"第三极"力量,在全球数字治理规则构建中发挥重要作用.生成式人工智能大模型全球治理应秉承"网络空间命运共同体"理念,先行设计合理的技术规制方案,利用《数字市场法》中的"守门人"制度,防范和化解模型运营商潜在的垄断风险,从技术角度将生成式人工智能大模型视为《数字服务法》中的"在线搜索引擎",为构建更具体的生成式人工智能大模型全球治理方案奠定基础.
The Theoretical Proof and Preliminary Conception of Global Governance of Generative Artificial Intelligence Foundation Models
As an important technical branch in the field of artificial intelligence,generative artificial intelligence not only plays a posi-tive role in driving the transformation of the digital economy and promoting the upgrading of industrial intelligence,but also causes a number of new social problems and national security risks,the proposal of global governance of generative artificial intelligence founda-tion models is timely.The EU is the"Third Pole"in the digital era after China and the United States,plays an important role in the con-struction of global digital governance rules.The global governance of the generative artificial intelligence foundation models should adhere to the concept of building a"Community of Common Destiny in Cberspace",design reasonable technical regulation schemes at first,use the"Gatekeeper"system in the Digital Market Act to prevent and resolve the underlying monopoly risk of model operators,and treat the generative artificial intelligence foundation models as"Online Search Engine"in the Digital Service Act from the technical perspective,lay the foundation for the future construction of a more specific and refined generative artificial intelligence foundation models global governance scheme.

Generative artificial intelligenceGlobal governanceDigital hegemonyCommunity of common destiny in cyberspaceSearch engine

严驰

展开 >

同济大学法学院,上海 200092

生成式人工智能 全球治理 数字霸权 网络空间命运共同体 搜索引擎

教育部人文社会科学研究规划基金湖北省知识产权局软科学研究计划(2023)河南省法学会民法学研究会研究课题(2024)

22YJA82003020232S0064HNCLS2024083

2024

中国科技论坛
中国科学技术发展战略研究院

中国科技论坛

CSTPCDCSSCICHSSCD北大核心
影响因子:1.081
ISSN:1002-6711
年,卷(期):2024.(5)
  • 52