Bar chart
AI answers with significant sourcing issues, by assistant
A bar chart from the EBU and BBC study News Integrity in AI Assistants, in which journalists from 22 public service media organisations rated AI answers to 3,113 news questions. It shows the share of each assistant's responses with significant sourcing issues, against 31% across all responses. The study covered news rather than products, but the same mechanism is how assistants misattribute claims to brands.
Where this appears
What's in this bar chart
| Significant sourcing issues | |
|---|---|
| Gemini | 72% |
| ChatGPT | 24% |
| Perplexity | 15% |
| Copilot | 15% |
Reference line: All responses, 31%.
Professional journalists rated answers to 3,113 news questions in 14 languages.
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/ai-assistant-sourcing-errors-by-assistant-ebu-bbc"><img src="https://www.koji.so/visuals/ai-assistant-sourcing-errors-by-assistant-ebu-bbc/ai-assistant-sourcing-errors-by-assistant-ebu-bbc.png" alt="Bar chart of AI answers with significant sourcing issues: Gemini 72%, ChatGPT 24%, Perplexity 15% and Copilot 15%, against 31% across all responses." width="1200" height="630" /></a>
<p>Source: <a href="https://www.koji.so/visuals/ai-assistant-sourcing-errors-by-assistant-ebu-bbc">AI answers with significant sourcing issues, by assistant</a>, Koji</p>
