首页|生成式人工智能的理论审视、风险揭示及治理路径——以ChatGPT为视角

生成式人工智能的理论审视、风险揭示及治理路径——以ChatGPT为视角

扫码查看
ChatGPT是生成式人工智能的典型代表,它具有卓越的语言理解和生成、自适应学习和个性化响应、多模态处理等能力.聚焦ChatGPT,深入探讨生成式人工智能的演进趋势,并剖析其在应用中可能引发的虚假信息泛滥、个人隐私泄露、信息安全及民众失业等社会风险,提出应从法律规制、伦理准则和技术路径等出发,明确政府、社会和个体的责任义务,完善生成式人工智能的法律监管体系,加强行业自律与标准制定,构建协同共治、动态适应的格局,以确保生成式人工智能技术的健康与可持续发展.
Theoretical review,risk disclosure,and governance path of generative artificial intelligence:A ChatGPT perspective
ChatGPT is a typical representative of generative artificial intelligence,which has excellent abilities in language understanding and generation,adaptive learning and personalized response,multimodal processing,etc.Focusing on ChatGPT,we will delve into the evolutionary trends of generative artificial intelligence and analyze the potential social risks it may pose in its applications,including the proliferation of false information,personal privacy breaches,information security,and unemployment among the general public.In view of this,we should clarify the responsibilities and obligations of the government,society,and individuals from the perspectives of legal regulation,ethical standards,and technological paths,improve the legal regulatory system for generative artificial intelligence,strengthen industry self-discipline and standard setting,and create a collaborative governance and dynamic adaptation pattern to ensure the healthy and sustainable development of generative artificial intelligence technologies.

ChatGPTgenerative artificial intelligencedate securityalgorithmic biassocial risks

胡枫、张超

展开 >

南京邮电大学 通信与信息工程学院,江苏 南京 210003

南京邮电大学 党委办公室,江苏 南京 210023

ChatGPT 生成式人工智能 数据安全 算法偏见 社会风险

2024

南京邮电大学学报(社会科学版)
南京邮电大学

南京邮电大学学报(社会科学版)

CHSSCD
影响因子:0.477
ISSN:1673-5420
年,卷(期):2024.26(6)