The AI & Global Health Hub

Stop Chasing the Perfect Prompt

Just talk to your AI

Claude Fable 5: How to use the smartest model without burning through your usage

Claude Skill: Token-Efficient Use

The AI you use for research reads about half the pages it claims to

Claude skill for full-page web search

Five AI Advisors Reviewed My Essay. They Found Five Different Problems.

Claude skill: LLM Council

Why Evaluating AI Health Interventions Is Broken (And What to Do About It)

I built a citation fact-checker. Here's how it compared to ChatGPT, Perplexity, and Claude

Claude skill: Citation verification

Is Your Data Safe When You Use AI?

Even if you opted out of training, it might not be.

AI is entering CHW programs faster than health systems can absorb it

Lancet Primary Care article on AI in CHW programs

AI doesn't just "hallucinate." Here are six types of errors I found while researching 37 health programs.

The word "hallucination" treats six different problems as one. Here's what actually goes wrong when AI does research, and what to do about each one.

The Dangers of Using AI to Do Research

AI Tools Are Amazing, but We Still Need Roads, Internet, and Trained Health Workers