This project analyzes misinformation based on the risk of harm it can cause, not just whether it is true or false. The system uses reasoning-based AI to study emotional tone, calls to action, and context, and then generates a Harm Index that explains the potential real-world impact of viral content.