智能化背景下的算法信任  

Algorithm Trust under the Background of Intelligence

作　　者：闫宏秀[1] 宋胜男 YAN Hong-xiu;SONG Sheng-nan(Institute of History and Culture of Science,Shanghai Jiao Tong University,Shanghai 200240,China)

机构地区：[1]上海交通大学科学史与科学文化研究院,上海200240

出　　处：《长沙理工大学学报（社会科学版）》2020年第6期1-9,共9页Journal of Changsha University of Science and Technology：Social Science

基　　金：国家社会科学基金规划项目(19BZX043)。

摘　　要：算法信任是技术信任的一种,隶属于传统意义上的人与技术之间的信任关系。算法与数据、智能的合力将这种技术信任推向了前所未有的境界。在数据被作为一种新型的生产要素的当下,作为数据核心动力之一的算法因其不透明性、偏见等技术不确定性问题所引发的算法风险构成了对算法信任的挑战。因此,需要从算法信任的缘起及其所蕴含风险的解析中找寻出路。有效的算法信任应当是算法自身的技术的鲁棒性与人类理性的契合。Algorithmic trust is a kind of technical trusts,which belongs to the traditional trust relationship between people and technology.The join force from algorithm,data and intelligence promotes the technological trust to an unprecedented level.In case that data is regarded as a new type of productive factors,algorithm,one of the core driving forces of data,has led to algorithmic risks that pose a challenge to algorithmic trust due to such technical uncertainties as opacity,bias,etc.Therefore,it is necessary to find a way out from analyzing the causes of algorithmic trust and the risks involved.The effective algorithmic trust is supposed to be the combination between algorithm's own technical robustness and human rationality.

关 键 词：算法 算法信任 调节 伦理 算法风险 算法设计 

分 类 号：TP391.3[自动化与计算机技术—计算机应用技术]