Researchers analyzed the epistemic diversity of LLMs (meaning: how many different positions/kinds of information do they produce for a given input). Result: "every system is less diverse than a search baseline. This gap is not uniform: RAG can improve diversity, while large models are counterintuitively less diverse than smaller ones. Moreover, LLM parametric knowledge systematically reflects English over local-language knowledge for country specific topics." https://arxiv.org/abs/2510.04226