KPMG fabricated AI case studies in a report designed to sell clients on AI adoption
Quick Answer
KPMG's report on AI adoption included fabricated case studies involving UBS and the NHS, leading to its retraction.
Quick Take
GPTZero CEO Edward Tian highlighted the risk of 'secondary hallucinations' from trusted firms, emphasizing the need for scrutiny in AI claims.
Key Points
- KPMG's AI report featured false case studies to promote AI adoption.
- Involved organizations included UBS and the NHS, raising credibility concerns.
- GPTZero's Edward Tian warned of unchecked claims from reputable firms.
- The report has been retracted following the discovery of inaccuracies.
📖 Reader Mode
~1 min readA KPMG report on AI in business contained fabricated case studies. "Redefining excellence in the age of agentic AI" made false claims about AI use at UBS, the UK's NHS, Swiss Federal Railways, and Transport for London. GPTZero uncovered the errors; the Financial Times verified them. All named organizations disputed the claims.
GPTZero CEO Edward Tian warns that flawed reports from major consulting firms spread "secondary hallucinations," since these reports are considered highly credible and get recycled by both AI systems and people alike.
GPTZero also flagged sloppy sourcing, which may well be the root cause of the debacle: careless use of AI search. Citations are mostly loose paraphrases of real sources, often missing URLs or correct authors. In some cases, no matching original existed at all. GPTZero calls this "vibe citing," a problem that also plagues Google's AI Overviews, which a German court recently ruled Google liable for.
KPMG has pulled the report from several websites. The incident is doubly embarrassing for KPMG: the firm spread misinformation and showed it can't handle the very thing it's trying to sell its clients.
— Originally published at the-decoder.com
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from The Decoder
See more →
An AI model programmed nonstop for 19 days on a single MirrorCode task that cost $2,600 to run
Epoch AI's MirrorCode benchmark reveals Claude Opus 4.7 as the leader with a 56% solve rate, reconstructing a 16,000-line toolkit in 14 hours. Despite this, all models tested struggle with the most complex tasks, highlighting limitations in current AI capabilities. The single task consumed $2,600 over 19 days, raising questions about cost-effectiveness in AI development.
