审计中人工智能使用、任务复杂性与投资者感知的审计师责任——一项实验研究  

The Use of Artificial Intelligence in Auditing,Task Complexity and Investor’s Perception of Auditors’Responsibility--An experimental evidence

作　　者：张继勋[1] 李方晗 赵艺 卿水娟 Zhang Jixun;Li Fanghan;Zhao Yi Qing;Shuijuan

机构地区：[1]南开大学商学院,300071 [2]北京物资学院,101149

出　　处：《审计研究》2024年第5期62-74,共13页Auditing Research

基　　金：国家自然科学基金项目(项目批准号:72272080、71872090、72302123)的资助。

摘　　要：近年来,随着人工智能等信息技术的不断发展,审计中人工智能使用的影响日益受到关注。本文运用心理学理论和实验研究方法,分析和检验了审计中人工智能的使用和任务的复杂性对投资者感知的审计师责任的影响。研究发现,与人类审计师审计后出现重大错报相比,在使用人工智能审计后仍出现重大错报的情况下,投资者感知的审计师责任更高。研究还发现,对于复杂的审计任务,与人类审计师审计后出现重大错报相比,在使用人工智能审计后仍出现重大错报的情况下,投资者感知的审计师责任更高;但对于简单的审计任务,无论是人类审计师还是使用人工智能审计后仍出现重大错报,投资者感知的审计师责任没有显著差异。此外,研究还发现,审计中所使用的审计方法和任务复杂性共同影响投资者感知的审计师责任的机制。本文对审计理论和审计师实践都具有重要意义。In recent years,with the development of information technology,the impact of the use of artificial intelligence(AI)in auditing has received increasing attention.Based on psychological theory and experimental research methods,this paper examines the effect of the use of AI in auditing and task complexity on investor’s perception of auditors’responsibility.We find that when a material misstatement is discovered subsequently after the issuance of audit report,investors perceive auditors to be more responsible for using AI to perform audit tasks.We also find that for complex audit tasks,when a material misstatement is discovered after audit,compared with human auditors performing audit tasks,investors’perception of auditors’responsibility is still higher when auditors use AI to perform audit tasks.However,for simple audit tasks,whether it is performed by human auditors or by AI,there is no significant difference in investors’perception of auditors’responsibility.In addition,we find that both audit methodology and the complexity of audit tasks affect investors’perception of auditors’responsibility.This paper has significant implications for audit theory and practice.

关 键 词：人工智能 人类审计师 任务复杂性 投资者 审计责任 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] F239.1[自动化与计算机技术—控制科学与工程]