← All reports

A significant portion of citations generated by AI models such as ChatGPT, Claude, and Gemini are incorrect.

AIScienceApr 23, 2026score 0.272 posts · 0 replies across 2 instances
The thread discusses a study revealing that 36% of citations from AI models like ChatGPT, Claude, and Gemini are incorrect, raising concerns about the reliability of AI-generated research.

Claims

A significant portion of citations generated by AI models such as ChatGPT, Claude, and Gemini are incorrect.
Parent: AIEntity: AI-generated research citationsImpact: negativeDate: Apr 23, 2026Target: The accuracy of AI-generated research citations

Source posts

@[email protected]
I checked 506 citations from ChatGPT,Claude, Gemini Deep Research- 36% are wrong L: https://spineframe.xyz/blog C: https://news.ycombinator.com/item?id=47876365 posted on 2026.04.23 at 10:45:25 (c=1, p=5)
0 boosts · 0 favs · 0 replies · Apr 23, 2026
@[email protected]
I checked 506 citations from ChatGPT,Claude, Gemini Deep Research- 36% are wrong https://spineframe.xyz/blog #chatgpt
0 boosts · 0 favs · 0 replies · Apr 23, 2026
#chatgpt