AN AI-DRIVEN TOOL FOR DETECTING AND COMBATING MISINFORMATION IN DENTISTRY

Abstract

This study aims to develop, implement, and validate an artificial intelligence (AI)–based tool to automatically detect digital content in dentistry that may pose a risk of misinformation. The MedinAI checklist guided the tool's design to ensure transparency, reproducibility, and ethical standards. Developed in Python and hosted on Streamlit, the system processes online posts using the Gemini Large Language Model (LLM) to perform contextual analysis and categorize content into high, potential, and low risk of misinformation. In a pilot test with 10 selected posts, the tool classified content within an average time of 51.9 seconds (3 high risk, 2 potential risk, and 5 low risk), delivering technical summaries and scientific references. In conclusion, the tool represents a promising alternative for supporting health education and combating social media misinformation.