In mid-June 2026, KPMG withdrew one of its reports on the use of artificial intelligence. The reason: the document contained data apparently generated by AI systems with evident hallucinations. Therefore, the published information was neither verifiable nor reliable.
This episode isn't an isolated incident. In fact, it's a systemic warning for all organizations using AI tools in their enterprise analyses. However, the problem doesn't just affect large consulting firms. Italian SMEs that rely on unsupervised AI outputs also risk building strategic decisions on flawed data. Consequently, governing AI-generated content becomes an operational priority, not just a theoretical concern.
We at SHM Studio We are closely following these developments. In particular, we are working with Italian SMEs to integrate AI into marketing and communication processes in a controlled and verifiable way. Finally, this KPMG case offers a concrete starting point for reviewing internal workflows and strengthening fact-checking processes for every artificially generated output.
The Timeline of the KPMG Case
On June 13, 2026, TechCrunch has reported that KPMG has withdrawn a report dedicated to the use of AI. The document, intended for an enterprise audience, contained unverifiable data and claims. The identified cause: hallucinations produced by the artificial intelligence systems used in its drafting.
The withdrawal happened quietly, but the news spread quickly in the industry. In fact, KPMG is one of the four major global consulting firms. Therefore, an error of this magnitude carries significant symbolic weight. This isn't a startup experimenting: this is an organization with structured internal processes.
However, the story doesn't surprise those who closely follow the evolution of language models. Generative AI systems tend to produce plausible outputs even in the absence of real data. This phenomenon, known as hallucination, has been documented for years in technical literature.
Winners and Losers in This Affair
Who is harmed by this episode? First and foremost, KPMG itself. The reputation of a consulting firm is based on the quality and accuracy of its analyses. A report withdrawn due to AI hallucinations weakens the trust of institutional clients.
Secondly, the entire ecosystem of enterprise AI vendors is affected. Many of these tools are sold as reliable solutions for generating reports and analyses. Consequently, episodes like this fuel skepticism among decision-makers. Among other things, some organizations might slow down their adoption plans.
Who can benefit from this situation, then? Companies and agencies that have established human oversight processes for AI output. Furthermore, professionals advocating a hybrid approach — AI as a tool, humans as validators — see their position strengthened. Similarly, AI governance frameworks now gain practical relevance that was more theoretical until yesterday.
Why Hallucinations Specifically Affect Analytical Reports
Large language models don't reason; they generate text that is statistically consistent with the prompt received. Therefore, when queried about specific data — percentages, statistics, research — they tend to produce plausible numbers even if no real sources support them.
This mechanism is particularly insidious in analytical reports. In fact, in a document of this type, a made-up but correctly formatted piece of data can pass superficial reviews. In particular, if the reviewer doesn't know the original source, checking becomes very difficult.
According to research by Gartner on Large Language Models , managing hallucinations remains one of the main challenges for the enterprise adoption of generative AI. Therefore, this is not a minor issue destined to disappear with future model updates.
Furthermore, the problem is amplified when AI is used to analyze data about AI itself. In this case, the models draw on a training corpus that may contain contradictory, obsolete, or AI-generated information. The result is a potentially self-referential error cycle.
SHM Studio's Take: A Governance Problem, Not a Technology One
We at SHM Studio we interpret the KPMG case as a process problem, not a tool problem. Generative AI is not inherently unreliable. However, it becomes dangerous when integrated into workflows lacking verification checkpoints.
The Right Question Isn't
Related articles
Discover more articles exploring similar topics, selected to offer you a more complete and stimulating perspective. Each piece of content is carefully chosen to enrich your experience.
