Linco Medical System · Posted Nov 14, 2026
Review two blinded AI-written Amharic patient instructions for the same clinical case, rate each for safety and language quality, and correct the exact words that are wrong.
Two leading AI models write Amharic patient instructions for the same clinical case. You review both side by side without knowing which model wrote which.
Your verdicts, written reasoning and word-level corrections become structured evidence showing where medical AI is unsafe, unclear or culturally wrong.
Spots per case are intentionally limited to protect review quality. Reviews are final once submitted, and no AI assistance is allowed while reviewing.
Sign in to your Linco dashboard and open AI Evaluation. The study opens after you read and accept the AI Evaluation Terms and Conditions.
Linco runs a preregistered, blinded study of how AI models write Amharic patient instructions.
Visit official website