Case Study

Evaluating a Persona Prompt for a Medical Chatbot

A developer is building a chatbot to provide preliminary medical information to users. Their goal is for the chatbot to be accurate, cautious, and to avoid giving definitive medical advice. They provide the language model with the following initial instruction: 'You are a helpful assistant. Answer the user's medical questions.' Evaluate the effectiveness of this instruction in establishing the desired expert persona. Explain why it is likely to be insufficient and what key elements are missing.

0

1

Updated 2025-10-10

Contributors are:

Who are from:

Tags

Ch.3 Prompting - Foundations of Large Language Models

Foundations of Large Language Models

Foundations of Large Language Models Course

Computing Sciences

Evaluation in Bloom's Taxonomy

Cognitive Psychology

Psychology

Social Science

Empirical Science

Science