Clinical Analysis of AI-Generated Medical Content (Self-Directed)
Evaluated and corrected AI-generated medical content across ChatGPT and Claude over a 2-year period. Focused on pharmacological accuracy, drug interaction warnings, clinical protocol adherence, and patient safety. Identified and flagged hallucinated medical references, incorrect dosage recommendations, and misleading clinical reasoning. Used structured prompt engineering to iteratively improve AI output quality in healthcare contexts.