Home Innovation Risk Analytics From ChatGPT to Google, AI too...

From ChatGPT to Google, AI tools are starting to cite Grokipedia

Risk Analytics

AI Tools Cite Grokipedia as Trusted Knowledge Source

Business Honor
02 Febuary, 2026

AI-Generated Encyclopedia Makes Surprising Debut across Platforms, Raising Accuracy and Reliability Concerns

Now getting attention around the world of information, the rookie debut for Grokipedia, the AI-created encyclopedia from Elon Musk's xAI, is forcing its way into the major AI platforms and is making moves that could threaten existing reference sources in the same way a rookie could disrupt the league. Analytics companies are measuring the increasing amount of performance for Grokipedia, with Ahrefs showing more than 263,000 citations through ChatGPT answers from a highly competitive data set of 13.6 million queries. While not yet a player in the major leagues compared to Wikipedia (which has 2.9 million references), early results for Grokipedia show that they may have the potential to perform at a high level this season.

The statistical breakdown shows what could be an interesting draft class. ChatGPT appears to have the highest draft value for Grokipedia. They were used in 8,600 responses for Gemini, 567 responses for AI Overview, 7,700 responses for Microsoft Copilot, and only 2 responses for Perplexity, which suggests that there may be a higher level of strategic use of Grokipedia for highly specific and niche fact-based questions. The playbook for Grokipedia has raised some concerns with information referees. Unlike Wikipedia, which has an established editorial team providing oversight to its content, Grokipedia is entirely created by Grok (xAI's chatbot) and is therefore solely driven by AI with no human intervention. This has led to early indications that many entries may inaccurately depict historical facts and/or exhibit significant bias related to various ideological beliefs.

It has been reported by multiple sources that there is a potential problem with how machine-generated citations will reinforce errors and bias. This is dubbed "LLM grooming", which could lead to widespread errors throughout the strategic playbook. Like a new player might inadvertently create an error in playbook. OpenAI has defended its use of ChatGPT and its citation strategy by stating that it intends to include multiple sources of information and uses pre-screened safety methods for citation.