KPMG Partner Fined Over AI Cheating
A partner at KPMG Australia was fined after staff were found to have used artificial intelligence tools to cheat on internal exams. The incident highlights emerging governance challenges for audit firms, placing renewed pressure on quality controls and independence as AI adoption accelerates.
- The partner, a registered company auditor, was fined over $10,000 and required to retake an internal training exam on the ethical use of AI after using a generative AI tool to answer questions. This individual self-reported the incident to Chartered Accountants Australia and New Zealand, which has initiated its own investigation. - This was not an isolated incident; KPMG Australia has identified 28 cases of AI-related misconduct within the current financial year, with the other 27 cases involving staff at or below the manager level. The firm began monitoring for AI use in internal testing in 2024 and plans to disclose the total number of AI-related cheating cases in its annual results. - This event follows a history of cheating scandals at the firm. In 2021, the US Public Company Accounting Oversight Board (PCAOB) fined KPMG Australia $450,000 for widespread answer sharing on internal training tests between 2016 and 2020, which involved over 1,100 employees, including partners. - The incident has drawn criticism from Australian politicians, with Greens senator Barbara Pocock labeling the oversight system as "toothless." The Australian Securities and Investments Commission (ASIC) has been notified but is waiting for the professional accounting body's disciplinary proceedings before taking further action. - Other major consulting firms are also grappling with AI governance. In 2025, Deloitte Australia refunded a portion of a $440,000 government contract after a report it produced using generative AI was found to contain fabricated quotes and references to non-existent research. - The episode highlights the broader challenge for audit and consulting firms in establishing clear governance and ethical guidelines for AI use as they simultaneously market their own AI-based consulting services. Regulators are still developing standards for AI in financial reporting and auditing, creating a complex compliance landscape. - For internal audit functions, the proliferation of AI necessitates new risk assessment protocols. Key areas of concern include data privacy, potential for algorithmic bias in financial analysis, and ensuring the explainability of AI-driven decisions to stakeholders and regulators.
