KPMG Report on AI Plagued by Hallucinations and Fake Citations

In October last year, KPMG published a report titled Total Experience: Redefining Excellence in the Age of Agentic AI, which was about how companies are using AI to cater to customers' needs. KPMG is one of the "Big Four" professional services and accounting firms in the world, along with Deloitte, PricewaterhouseCoopers and Ernst & Young.

Apparently, though, that report was full of AI hallucinations and included examples of agentic AIs that either did not exist or did not have the capabilities KPMG stated in the paper. Investigators for GPTZero, the maker of an AI content detection tool, found inaccuracies and fake footnotes all over the report, which were also verified by the Financial Times.

In its report of the investigation, GPTZero said that only five citations out of 45 in the paper accurately pointed to real sources. A total of 28 citations paraphrased titles or added fake components to real sources, while 12 were phrased too vaguely to determine whether they actually existed. GPTZero called the creation of fake references by AI models "vibe citing."

In addition to the fake or inaccurate citations, the investigators also found that approximately half of the claims in the paper were fake or misattributed. They were "likely the result of an AI research tool over-complying with a request to find examples of 'agentic AI' in the wild," GPTZero wrote. In one example, KPMG claimed that Emirates launched a mobile chatbot called Sara that can talk to passengers and alter their flights for them. Sara was a mobile assistant launched in 2023 and not an AI-powered chatbot, and it also didn't have the power to change bookings for passengers.

KPMG also claimed that Swiss multinational investment bank UBS integrated agentic AI across its "investment advisory, risk management and compliance monitoring." The bank told the Times that the information was "factually incorrect." In another example, KMPG said that Swiss Federal Railways (SBB) has AI agents that can help passengers plan, book and optimize their trips based on preferences, real-time conditions and carbon impact. An SBB spokesperson said that was "not accurate."

Papers by companies like KPMG are typically cited in other research papers and articles, since they're considered as highly trusted sources. GPTZero chief executive Edward Tian explained that error-riddled papers published by the Big Four could "poison the well of information" and could lead to second-hand AI hallucinations. A KPMG spokesperson told the Times that the company "takes the accuracy and integrity of its published content seriously." KPMG has since pulled the paper and is now "reviewing the circumstances surrounding its publication."