【摘要】 人工智能技术在审计领域的深入应用，在提升效能的同时也引发了前所未有的伦理挑战。本文系统识别、剖析了AI审计的算法与数据偏见、责任边界模糊、不透明的“黑箱”效应与数据安全与隐私问题四大核心伦理风险，并建构了以“价值伦理为引领、过程控制为支柱、算法治理为核心”的三维治理框架，旨在将伦理要求内嵌于AI审计实践中。最后，本文建议多主体协同、分三阶段逐步构建可信赖的智能审计生态体系。更多还原

【Abstract】 With the implementation of the new Securities Law and The in-depth application of artificial intelligence technology in the audit field has improved efficiency and also brought about unprecedented ethical challenges.This study systematically identifies and analyzes four core ethical risks of AI auditing,namely algorithmic and data bias,blurred responsibility boundaries,opaque "black box" effect,and data security and privacy issues.It also constructs a threedimensional governance framework featuring "value ethics as the guidance,process control as the pillar,and algorithm governance as the core",aiming to embed ethical requirements into the practice of AI auditing.Finally,this paper proposes to gradually build a trustworthy intelligent audit ecosystem through multistakeholder collaboration in three stages.更多还原