Archive — US — techcrunch
KPMG pulls report on AI usage due to apparent hallucinations
KPMG has retracted a report on AI usage due to inaccuracies attributed to AI hallucinations, following claims by various organizations that the report misrepresented their AI practices. The firm is investigating the matter while affirming its commitment to responsible AI use and oversight.
Key points
- KPMG published a report on AI usage in October 2025.
- Numerous organizations, including UBS and the NHS, claimed the report contained false information.
- GPTZero identified inaccuracies in the report related to AI hallucinations.
- KPMG removed the report from its website pending an investigation.
- The firm emphasizes the importance of human oversight in AI applications.
Why it matters
- AI hallucinations are a known issue impacting the reliability of AI-generated content.
- This incident reflects ongoing challenges in integrating AI responsibly, highlighted by similar problems faced by other firms like EY, which previously withdrew a report for inaccuracies.
- Given the rise of AI in business applications, ensuring accuracy and accountability in AI outputs remains critical.
KPMG's retraction underscores the pressing need for rigorous oversight in AI use to avoid misinformation, reflecting broader industry challenges and the necessity for responsible AI deployment in future projects.