The Application of Artificial Intelligence in Auditing Sustainability Reports: Between Ethical Challenges and Efficiency

Abstract

Artificial intelligence is increasingly embedded in the assurance of sustainability reports, promising efficiency gains yet raising acute ethical concerns. This study examines how AI can improve audit efficiency while preserving ethical integrity, and whether such use remains compliant with emerging assurance and reporting standards. Building on a three-pillar model: Ethics and Independence, Efficiency and Detection, and Governance and Compliance, we derive three hypotheses and evaluate them through a practitioner survey of auditors, accountants and senior managers, complemented by a targeted document analysis using LLM/NLP tools on ESG reports. Results indicate that explicit ethical safeguards (bias testing, transparency and independence protocols) are associated with higher stakeholder trust; moreover, AI-assisted review reduces time while maintaining or improving the detection of exaggeration/greenwashing; finally, aligning AI workflows with ISSA 5000, IFRS S1/S2, ESRS and GRI enhances traceability of evidence and perceived compliance. Accordingly, the study underscores the need for continuous professional training, clear ethical guidelines and governance-by-design for AI in sustainability assurance, and offers an integrated model that connects ethics, efficiency and standard-based compliance. JEL: M42; M41; M14; Q56; C45