KPMG Retracts AI Report Following Reports of Factual Inaccuracies

Professional services giant KPMG has officially retracted its recent report, "Redefining excellence in the age of agentic AI," following public pushback from several major organizations. The report, which aimed to analyze the current state of AI adoption, was pulled after entities including the UK’s National Health Service, UBS, Swiss Federal Railways, and Transport for London disputed the accuracy of the claims made regarding their specific AI implementations. Research firm GPTZero identified these errors as AI hallucinations, suggesting that KPMG utilized generative AI tools to draft the report without sufficient human verification.

This incident highlights a growing trend of institutional reliance on automated content generation, often at the expense of rigorous fact-checking. By using AI to write a report about the efficacy of AI, KPMG inadvertently demonstrated the very risks—unreliable outputs and lack of source verification—that the technology is often criticized for. The firm has since launched an internal investigation, emphasizing that its internal guidelines require human oversight to validate content and verify independent sources, a protocol that clearly failed in this instance.

This retraction is part of a broader pattern of professional services firms struggling with the integration of generative AI into their research workflows. Notably, EY recently withdrew a similar report on loyalty programs after discovering fabricated footnotes and AI-generated hallucinations. These events serve as a cautionary tale for the corporate sector: while AI can accelerate content production, the absence of robust human-in-the-loop validation processes can lead to significant reputational damage and the dissemination of misinformation. As firms continue to adopt these tools, the industry will likely face increased pressure to implement stricter governance frameworks to ensure the integrity of their published research.