A senior European journalist has been suspended for publishing artificial intelligence-generated quotes presented as real.
Peter Vandermeersch inserted dozens of fabricated quotes into 15 of 53 articles he wrote for two Mediahuis websites. An investigation by Dutch newspaper NRC found the quotes did not appear in the cited sources-news reports or scientific studies-and seven individuals confirmed they never made the attributed statements.
Vandermeersch, former CEO of Mediahuis Ireland (2022-2025) and currently a journalism fellow with the European publishing group, admitted the misconduct on his Substack blog. He said he used AI tools-including ChatGPT, Perplexity, and Google Notebook-to summarize lengthy documents, mistakenly trusting their outputs.
"That was not just careless, it was wrong," he wrote, acknowledging he committed the very error he had warned colleagues against: relying on AI's convincingly realistic but false quotations.
He first discovered the problem last year but failed to correct the articles immediately. While he still uses AI for translation and idea generation, he now does so with far greater caution.