从算法黑箱到算法透明:政府算法治理的转轨逻辑与路径  被引量:27

From Algorithm Black Box to Algorithm Transparency: the Transition Logic and Path of Algorithm Governance by the Government

在线阅读下载全文

作  者:张红春 章知连 ZHANG Hongchun;ZHANG Zhilian(School of Public Administration,Guizhou University,Guiyang,Guizhou,China,550025)

机构地区:[1]贵州大学公共管理学院,贵州贵阳550025

出  处:《贵州大学学报(社会科学版)》2022年第4期65-74,共10页Journal of Guizhou University(Social Sciences)

基  金:国家社会科学基金青年项目“地方政府公务员大数据素养的结构、测度与培育路径研究”(21CZZ035)。

摘  要:在大数据和人工智能时代,依循算法的政府算法治理模式日渐兴起,它让政府治理日益自动化、智能化和精准化,但也带来了挥之不去的算法黑箱问题。算法黑箱缘何形成,又如何破解,这是政府算法治理必须解决的重大理论与实践问题。研究发现,政府算法的低透明度是算法技术自身特质和政府使用算法技术的制度方式耦合的产物,与政府治理的公共性、公开性、有效性诉求本质悖逆,必然要求从算法黑箱向算法透明转轨。其转轨的方向是构建算法决策、研发、应用、监督全过程透明化的政府算法治理体系,在其实现路径上则依赖政府与公众各自的透明责任与行动义务的双向构建。In the era of big data and artificial intelligence, the algorithm-based government governance model is rising.It makes government governance increasingly automated, intelligent and precise, but it also brings major theoretical and practical problems.The source of algorithm black box as well as its solution is of significant as theoretical and practical problems of government governance.This study found that the low transparency of government algorithms is the result of the characteristics of algorithm technology itself combining with the institutional way the government uses algorithm technology, which is the opposite of the essence of the requirements of publicity, openness and effectiveness, and therefore, the transition to algorithm is necessary.The direction of its transition is to build a fully transparent government algorithm governance system, which is composed of algorithm decision-making, research and development, application, and supervision.Its realization path relies on the two-way construction of transparency responsibility and action obligation by the government and the public respectively.

关 键 词:算法治理 算法透明 算法公开 算法参与 透明政府 人工智能 大数据 

分 类 号:C93-03[经济管理—管理学]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象