摘要
随着人类与人工智能的深度交融,人类与人工智能的关系具有了超越工具主义的内在特征。在此背景下,可信已被确立为引导人工智能未来发展的规范性要求,发展可信人工智能已成为全球性的共识。有鉴于“人—机”关系与人际关系的差异,发展可信人工智能的关键在于增强人工智能的可信任性,而不是谋求人工智能对“类人信任”前置条件的满足。发展可信人工智能需要在认识论上立足于存在论的现象学视域;在方法论上实现由个人契约论方法向社会现象学方法的转变;在现实性上则需要通过构筑社会技术系统的信任网络方能最终增进对人工智能的信任。
With the deep integration of human beings and Artificial Intelligence(AI),the relationship between human beings and AI has inherent characteristics beyond instrumentalism.In this context,“trustworthiness”has been established as a normative requirement to guide the future development of AI,and the development of trustworthy AI has become a global consensus.Given the differences between“human-machine”relationships and interpersonal relationships,the key to developing trustworthy AI lies in enhancing the trustworthiness of artificial intelligence,rather than seeking AI to meet the preconditions of“anthropomorphic trust”.The development of trustworthy AI should be based on the phenomenological perspective of ontology in epistemology.In terms of methodology,the transformation from contractarian-individualist approach to social phenomenological approach should be realized.In reality,it is necessary to build a trust network of sociotechnical system to finally increase the trust of artificial intelligence.
作者
刘永安
LIU Yong'an(School of Law,Nanjing Normal University Taizhou College,Taizhou 225300,Jiangsu,China)
出处
《昆明理工大学学报（社会科学版）》
2023年第4期27-33,共7页
Journal of Kunming University of Science and Technology(Social Sciences)
基金
教育部人文社会科学研究规划项目“前瞻性治理视域下新兴技术伦理评估的理论和方法研究”(22Y JAZH071)
江苏高校哲学社会科学研究项目“人工智能体的道德责任问题研究”(2021SJA2474)。
关键词
可信人工智能
可信任性
人机共生
社会现象学方法
社会技术系统
Trustworthy AI
trustworthiness
human-machine symbiosis
social phenomenological approach
sociotechnical system
