# When Cow Urine Cures Constipation on YouTube: Limits of LLMs in Detecting Culture-specific Health Misinformation

arXiv: https://arxiv.org/abs/2604.22002 · HTML: https://mlsanity.com/paper/2604.22002/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 51 references that are in the map.
- First posted on arXiv 23 Apr 2026.

## Builds on (most cited first, 8 of 8)

- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formatting](https://mlsanity.com/paper/2310.11324/) (2023)
- [Survey of Cultural Awareness in Language Models: Text and Beyond](https://mlsanity.com/paper/2411.00860/) (2024)
- [The Order Effect: Investigating Prompt Sensitivity to Input Order in LLMs](https://mlsanity.com/paper/2502.04134/) (2025)
- [HEALTH-PARIKSHA: Assessing RAG Models for Health Chatbots in Real-World Multilingual Settings](https://mlsanity.com/paper/2410.13671/) (2024)
- [Through the Prism of Culture: Evaluating LLMs' Understanding of Indian Subcultures and Traditions](https://mlsanity.com/paper/2501.16748/) (2025)
- [Lost in Translation? A Comparative Study on the Cross-Lingual Transfer of Composite Harms](https://mlsanity.com/paper/2602.07963/) (2026)
- [How much does context affect the accuracy of AI health advice?](https://mlsanity.com/paper/2504.18310/) (2025)

## Abstract (arXiv)

Social media platforms have become primary channels for health information in the Global South. Using gomutra (cow urine) discourse on YouTube in India as a case study, we present a post-facto Large Language Model (LLM)-assisted discourse analysis of 30 multilingual transcripts showing that promotional content blends sacred traditional language with pseudo-scientific claims in ways that sophisticated debunking content itself mirrors, creating a rhetorical register that LLMs, trained predominantly on Western corpora, are systematically ill-equipped to analyse. Varying prompt tone across three LLMs (GPT-4o, Gemini 2.5 Pro, DeepSeek-V3.1), we find that culturally embedded health misinformation does not look like ordinary misinformation, and this cultural obfuscation extends to gendered rhetoric and prompt design, compounding analytical unreliability. Our findings argue that cultural competency in LLM-assisted discourse analysis cannot be retrofitted through prompt engineering alone.