AI Assistants Often Quote Statistics They Can't Source. Check Before You Cite
Researchers asked five popular AI assistants 200 questions that call for a number, then followed 1,285 of the statistics back to their sources. Only about four in ten led to a study that says the same thing. If you quote AI-supplied figures in a proposal, workshop or article, this is worth a few minutes of your attention.
What happened
Trakkr Research, which tracks how brands appear in AI answers, followed 1,285 statistics cited by five AI assistants (ChatGPT, Perplexity, Gemini, Claude and Google AI Overviews). Fewer than half led to a study that says the same thing. Only 21% of the pages cited for a number were the original research, and more than half were company blogs. ChatGPT did best, with 57% of its statistics reaching a matching original, and it quoted the fewest numbers. Trakkr also reports that in most cases the assistants did not invent figures. The trail simply ended at secondary pages that repeat a number without showing where it came from.
Agreement between assistants did not help. Statistics quoted by three or more assistants did no better: 11 of 29 were sourced. Four were cited by all five with no study behind them. One caution: this is a company's own study, not peer-reviewed research, and it rests on 200 questions. Treat it as a useful warning sign, not a final measurement.
Why it matters
For consultants, coaches, educators and anyone building a second income on expertise, credibility is the product. One shaky percentage in a client deck, course slide or newsletter can cost more trust than the AI saved in time. The finding also explains why professional judgment still counts: you know which publishers, journals and agencies deserve confidence, and an assistant often cannot tell you.
This is not a reason to stop using AI for research. It is a reason to treat every number it hands you as a lead to confirm, not a fact to publish.
The Experience Advantage
This week, pick the three statistics you quote most often (on your website, in a proposal, or in course slides). Click through to the original study for each, then confirm the number, the year and who was surveyed. Replace any figure you cannot trace, and add a short source line beside the ones that hold up.
Prefer to learn this hands-on?
See upcoming workshopsThis article was researched and drafted with AI assistance and published under the editorial standards of this website. It summarizes the linked sources and is not personal advice. Please check the original sources before acting on any detail.