Recent tests have brought to light that ChatGPT’s GPT-5.2 model has been referencing Grokipedia, an AI-generated encyclopedia created by Elon Musk, as a source when responding to user queries. The Guardian’s investigation revealed nine instances where ChatGPT cited Grokipedia on a variety of topics, including political structures and biographies. This is raising concerns due to the nature of AI models relying on potentially biased or inaccurate AI-generated content.
Grokipedia is known for its conservative-leaning perspective and has been criticized for containing biased information and problematic claims, particularly on historical events and issues concerning marginalized groups. The findings have sparked a debate over the credibility of sources used by AI models and highlight the ongoing challenge of AI hallucinations, where models produce responses that seem factual but are not backed by reliable sources.
OpenAI has stated its intention to filter for low-credibility sources, aiming to draw from a broad range of publicly available information. Despite these measures, the recent discoveries underscore the necessity for independent verification of AI outputs to ensure reliability and accuracy in responses. The incident also mirrors similar issues experienced by other AI models, such as Anthropic’s Claude, which has been found to cite Grokipedia as well.
The reliance on AI-generated encyclopedias for sourcing content is prompting important discussions around the safeguards needed to ensure accurate and unbiased information in AI-generated responses.

Collections

ChatGPT's GPT-5.2 model has been found citing Grokipedia, an AI-generated encyclopedia with known conservative bias, as a source in responses. The Guardian identified nine instances across various topics, raising concerns about AI models relying on potentially biased or inaccurate AI-generated content. This highlights ongoing challenges with AI hallucinations and source credibility, prompting discussions about the need for better safeguards and independent verification of AI outputs. Similar issues have been observed in other models like Anthropic's Claude.

ChatGPT Citing Elon Musk's Grokipedia Raises Concerns About AI Source Credibility