摘要
随着生成式人工智能对社会的全面赋能,人工智能生成的信息也大量涌入社会各领域。虽然这满足了社会对信息的需求,保证了信息的时效性,但是,生成式人工智能的算法缺陷以及用户自身认知局限使虚假信息层出不穷;同时,在深度学习算法的基础上,虚假信息也达到了以假乱真的程度,给虚假信息的现有法律监管体系带来严峻考验。针对人工智能助推的虚假信息肆虐的现象,我们在重新审视虚假信息的概念框架与探析虚假信息既存法律规定的基础上,分析生成式人工智能视角下虚假信息治理的困境。通过借鉴欧盟对虚假信息的治理思路,提出虚假信息的治理建议:多元化主体相结合的社会共同治理、追根溯源的类型化治理以及建立多维度责任追究机制。人工智能技术是不断发展和革新的,生成式人工智能视域下虚假信息的治理是一项持续而复杂的工作,我国应加快完善人工智能领域的法律法规,同时亦需要法律界、科技界以及产业的共同努力,以降低人工智能的负面效应,优化其社会效益。
With the comprehensive empowerment of the society by generative AI,the information generated by AI has also flooded into various fields of society.Although this satisfies the society's demand for information and ensures the timeliness of information,the algorithm defects of generative AI and the user's own cognitive limitations make disinformation emerge in an endless stream.At the same time,on the basis of deep learning algorithms,false information has also become indistinguishable from the real one,which brings a severe test to the existing legal supervision system of false information.In view of the rampant phenomenon of disinformation promoted by AI,we re-examine the conceptual framework of false information and analyze the existing legal provisions of false information,and analyze the dilemma of false information governance from the perspective of generative AI.Based on the European Union's ideas on the governance of false information,this paper puts forward some suggestions on the governance of false information:social joint governance with the combination of diversified subjects,typological governance to trace the root and the establishment of a multi-dimensional accountability mechanism.AI technology is constantly developing and innovating,and the governance of false information from the perspective of generative artificial intelligence is a continuous and complex work.China should speed up the improvement of laws and regulations in the field of AI,and it also needs the joint efforts of legal,scientific and technological circles and industry to reduce the negative effects of AI and optimize its social benefits.
作者
黎明
张益欣
LI Ming;ZHANG Yixin(School of Law,Lanzhou University of Finance and Economics,Lanzhou 730010,Gansu,China)
出处
《昆明理工大学学报(社会科学版)》
2025年第1期1-9,共9页
Journal of Kunming University of Science and Technology(Social Sciences)
关键词
虚假信息
生成式人工智能
多元化治理
类型化治理
多维度责任分配
disinformation
generative AI
diversified governance
typological governance
multidimensional assignment of responsibility
