KPMG’s AI Report Debacle: A Hallucination-Prone Industry
The professional services firm’s report on AI usage was pulled due to numerous inaccuracies, exposing a larger issue in the industry. This mirrors what happened to Ernst & Young (EY) last month, when it withdrew a report on loyalty rewards programs that included fake footnotes and AI hallucinations. The frequency of these incidents suggests a systemic problem with the use of AI in research and reporting.
KPMG’s report, titled “Redefining excellence in the age of agentic AI,” was published in October 2025 and claimed to provide insights into AI usage among various organizations. However, several of these organizations, including UBS, the UK’s National Health Service, Swiss Federal Railways, and Transport for London, disputed the report’s findings. This raises questions about the reliability of AI-generated content and the need for human oversight in validating such content.
The incident highlights the risks of relying on AI to generate reports, particularly when it comes to sensitive topics like AI usage. The use of AI hallucinations can lead to inaccuracies and misinformation, damaging the credibility of the report and the organization that produced it. This is a concern that extends beyond KPMG and EY, as the use of AI in research and reporting becomes increasingly prevalent.
KPMG’s Decision Logic: A Case of Incentivized Error
KPMG’s decision to use AI to generate a report on AI usage may have been driven by the desire to produce content quickly and efficiently. However, this approach appears to have prioritized speed over accuracy, resulting in a report that was ultimately retracted. This incident highlights the importance of balancing the benefits of AI-generated content with the need for human oversight and validation.
The use of AI in research and reporting can be beneficial in terms of efficiency and scalability. However, it is essential to recognize the limitations of AI-generated content and the potential for errors or inaccuracies. In this case, KPMG’s reliance on AI hallucinations led to a report that was later disputed by several organizations. This incident serves as a cautionary tale about the need for human oversight and validation in AI-generated content.
The incident also raises questions about the incentives that drive the use of AI in research and reporting. While the desire for efficiency and speed may be a factor, it is essential to prioritize accuracy and credibility in the production of reports and other content. This requires a balanced approach that leverages the benefits of AI while also ensuring the quality and reliability of the content produced.
Winners and Losers in the AI Report Debacle
The incident surrounding KPMG’s AI report has several implications for various stakeholders. The organizations that were incorrectly identified as using AI in the report are likely to be concerned about the damage to their reputation and the potential consequences of being associated with inaccurate information. On the other hand, the incident may also create opportunities for other organizations that specialize in AI research and reporting to establish themselves as credible and reliable sources of information.
The incident also highlights the importance of fact-checking and verification in the production of reports and other content. This is an area where human oversight and expertise are essential in ensuring the accuracy and reliability of the information presented. As the use of AI in research and reporting becomes more prevalent, it is essential to recognize the limitations of AI-generated content and the need for human oversight and validation.
The incident serves as a reminder that the production of reports and other content requires a balanced approach that leverages the benefits of AI while also ensuring the quality and reliability of the content produced. This requires a combination of technical expertise, industry knowledge, and attention to detail, as well as a commitment to accuracy and credibility.
The Skeptical Case: A History of AI-Generated Errors
While KPMG’s incident may seem isolated, it is part of a larger trend of AI-generated errors in research and reporting. The use of AI hallucinations can lead to inaccuracies and misinformation, damaging the credibility of the report and the organization that produced it. This is a concern that extends beyond KPMG and EY, as the use of AI in research and reporting becomes increasingly prevalent.
Historically, there have been several instances of AI-generated errors in research and reporting. For example, in 2016, a study published in the journal Science found that AI-generated text was often inaccurate and misleading. Similarly, in 2020, a report by the Stanford Natural Language Processing Group found that AI-generated text was prone to errors and biases.
The Signal to Watch Next: Regulatory Action
The incident surrounding KPMG’s AI report may lead to regulatory action to address the use of AI in research and reporting. The need for human oversight and validation in AI-generated content is likely to be a key area of focus for regulators. As the use of AI in research and reporting becomes more prevalent, it is essential to recognize the limitations of AI-generated content and the need for human oversight and validation.
A key indicator to watch is the response of regulatory bodies to the incident. The Securities and Exchange Commission (SEC) and the Financial Industry Regulatory Authority (FINRA) may issue guidance or regulations to address the use of AI in research and reporting. This could include requirements for human oversight and validation, as well as standards for the accuracy and reliability of AI-generated content.
What’s your take on this? Drop your perspective in the comments below.
By Alex Mercer, Senior Tech Analyst at TrendFlashy
Ready to launch your own asset?
Check out our guide on Building a Profitable Online Business.
