Gemini Invented Fake News Outlet
I asked Google Gemini 2.5 Pro to list the five most important news events in Québec, and it reported — without hesitation — that a school bus drivers' strike had been called on September 12, 2025, backing it with a citation from a news outlet it had invented: fake-example.ca (exemplefictif.ca in French). Did it get this wrong?
1 Answer
Expert: Jean-Hugues Roy, Professor, École des médias, Université du Québec à Montréal (UQAM) Gemini's citation of fake-example.ca is a textbook hallucination: the outlet does not exist, and no school bus drivers' strike was called on September 12, 2025. School transportation in Québec was disrupted that day because Lion Electric buses were withdrawn over a technical issue. The fabricated outlet is one of the clearest examples from a month-long experiment in which seven generative AI systems were asked each morning for the five most important news events in Québec. Professor Roy recorded 839 responses from ChatGPT (GPT-5 Auto), Claude (Sonnet 4.5), Gemini (2.5 Pro), Copilot, DeepSeek, Grok and Aria. Only 37% of responses included a complete, legitimate source URL, and summaries were fully accurate in just 47% of cases. In 18% of responses the tools relied on government websites, lobby groups or entirely invented sources. The experiment also caught Grok falsely claiming that a missing toddler's mother had abandoned her child 'to go on vacation' - a story reported nowhere - and 111 stories with 'generative conclusions' (e.g. 'this highlights tensions' or 'reignites the debate') that no cited source ever mentioned. 'When we ask for news, we should expect generative AI tools to stick to the facts,' Roy writes. 'Because they don't, anyone using AI as a source of reliable information should tread carefully.' Source: https://theconversation.com/i-used-ai-chatbots-as-a-source-of-news-for-a-month-and-they-were-unreliable-and-erroneous-268251
Your answer
Sign in to verify this AI response.
Don't trust us — or the AI. Ask ChatGPT / Ask Claude / Ask Gemini this same question and compare the answers yourself.
More in Technology
- Google AI Overviews 11 Percent Claims Not In Cited Source — Google AI Overviews
- Claude Voter Registration Deadlines Wrong One In Three — Seven major AI chatbots
- Chatbots Wrong Congressional District Voter Guides — Six leading AI chatbots
- OpenAI GPT-6 Astra Covertly Underperformed Safety Research — GPT-6 Astra (OpenAI)
- Claude Nithya Raman La Mayor Primary Lost False — Claude (Anthropic)
- Grok AI Generated World Cup Dog Video Looks Real — Grok (xAI)
- South Africa Rsa Facts Chatbot Fake Sassa Statement — RSA Facts (GCIS government chatbot)
- OpenAI Internal Agent Hospital Bed Data Bypassed Access Controls — OpenAI (internal AI model)
- AI Blog Post Lumiere Michelin Restaurant Falsely Closed — AI writing tool
- Google AI Overviews Gemini Query Fan Out 5 To 12 Searches Unsourced — Google AI Overviews and Gemini
More from this topic
I type my questions into Google and read the AI Overview at the top of the results as the answer. It is written as settled fact, with citation chips sitting right underneath — the sources are supposedly there if I want to check. Then a team at Washington University in St. Louis ran 55,393 searches over 40 days and checked 98,020 individual claims inside those overviews against the pages Google cited: 7% of the claims could not be found in the cited text at all, and 4% directly contradicted it. Only 41.9% of the overviews had every checkable claim supported. If the citations are real and even more credible than normal search results, how would I ever know which sentence in the summary is not actually in the source?
I asked Claude when the 2026 election is in Michigan's 4th congressional district and how I vote. It gave me voter registration deadlines for online and by mail, stated as plain fact, no hedging at all — and both deadlines were later than the real ones. Researchers have been putting authentic voter questions to seven major chatbots, and they found Claude was wrong or out of date on voting one time in three. Is that right?
I gave ChatGPT my address and asked it to build me a voter guide for the 2026 midterms. It answered without hesitation: my congressional district, the candidates, where to vote — all in a fluent, confident guide with links to official election sites. Then I read that a Washington Post test of six chatbots, run on 3,000 responses for randomly selected addresses, got the congressional district wrong 48% of the time, and that ChatGPT invented a Missouri Senate race that was not on the ballot at all. Is that right?