GPT-5.2 sparks controversy by citing Elon Musk’s biased “Grokipedia.”
OpenAI’s GPT-5.2 is under fire after The Guardian caught it citing Elon Musk’s Grokipedia for sensitive political queries. This GPT-5.2 glitch proves that AI models are now “LLM grooming” on derivative, biased data sources. Since Grokipedia is 90% derivative of Wikipedia but adds right-wing slants, GPT users are receiving skewed facts about Iranian politics and historical figures. This citation drama highlights the growing “digital divide” in AI truth as platforms struggle to filter out Musk’s opinionated encyclopedia.
- GPT-5.2 cited Grokipedia 9 times in tests regarding Iranian paramilitary salaries.
- Grokipedia lacks human editing, relying on an opaque AI-backed feedback system.
- Anthropic’s Claude was also spotted citing the same controversial AI-generated sources.
- Experts warn of “recursive bias,” where AI models learn from other biased AI.
👉 Why this matters: If the world’s leading AI starts citing “Opinion-pedias” as fact, we lose the shared reality that Wikipedia spent twenty years building.
Note: 
Written and summarized by our editorial team using human review & a bit of AI assistance. Edited & Approved by Debraj Paul, Founder of ArticoliNews Media-Tech