All frameworks and cheat sheets

Key figure

How often AI assistants get the news wrong

Stat card: 20% of AI assistant answers had significant accuracy problems across 3,113 questions in the EBU and BBC study.
Outdated facts were among the most common failures, and the answers never show their age. Source: EBU and BBC, News Integrity in AI Assistants, October 2025

The headline figure from News Integrity in AI Assistants, published by the European Broadcasting Union and the BBC in October 2025: across 3,113 questions in 14 languages, scored by professional journalists, 20% of AI assistant answers had significant accuracy problems. Outdated information was one of the most common causes. Useful for anyone explaining why brands should measure what buyers believe, not just what models say.

Where this appears

What's in this key figure

20% of AI assistant answers had significant accuracy problems in a study of 3,113 questionsJournalists from 22 public service media organisations in 18 countries scored the answers. Outdated information was one of the most common issues.

Use this image

Free to use in your docs, decks and posts under CC BY 4.0, as long as you credit Koji with a link to this page. To embed it, paste this into your page:

html
<a href="https://www.koji.so/visuals/ebu-bbc-ai-assistant-accuracy-stat"><img src="https://www.koji.so/visuals/ebu-bbc-ai-assistant-accuracy-stat/ebu-bbc-ai-assistant-accuracy-stat.png" alt="Stat card: 20% of AI assistant answers had significant accuracy problems across 3,113 questions in the EBU and BBC study." width="1200" height="630" /></a>
<p>Source: <a href="https://www.koji.so/visuals/ebu-bbc-ai-assistant-accuracy-stat">How often AI assistants get the news wrong</a>, Koji</p>