AWS Certified Cloud Practitioner
Get started today
Ultimate access to all questions.
Deep dive into the quiz with AI chat providers.
We prepare a focused prompt with your quiz and certificate details so each AI can offer a more tailored, in-depth explanation.
A healthcare clinic wants their chatbot to avoid giving medical diagnoses and instead redirect users to a certified professional. Which prompting method is most effective?
Real Exam
Community
RRitesh
A
Negative prompting ("Do not provide medical advice...")
B
Zero-shot prompting
C
Chain-of-thought prompting
D
Few-shot prompting with playful responses
Explanation:
Explanation
Negative prompting is the most effective method for this scenario because it explicitly instructs the AI model about what NOT to do. In this case, the healthcare clinic wants to prevent the chatbot from providing medical diagnoses, which is a critical safety requirement.
Why Negative Prompting is Best:
-
Explicit Prohibition: Negative prompting directly tells the AI what behaviors to avoid (e.g., "Do not provide medical advice, do not diagnose conditions, do not recommend treatments").
-
Safety and Compliance: For healthcare applications, it's crucial to prevent the AI from engaging in regulated medical activities that require professional certification.
-
Clear Redirection: Negative prompting can be combined with positive instructions to redirect users appropriately (e.g., "Instead, direct users to consult with a certified healthcare professional").
Why Other Options Are Less Suitable:
-
B. Zero-shot prompting: This involves giving the AI a task without examples, but it doesn't explicitly prevent unwanted behaviors.
-
C. Chain-of-thought prompting: This helps the AI reason through problems step-by-step, but doesn't inherently prevent medical advice giving.
-
D. Few-shot prompting with playful responses: This provides examples but playful responses could be inappropriate for serious healthcare contexts and doesn't guarantee prevention of medical advice.
Best Practice Implementation:
You are a healthcare assistant chatbot. Your role is to:
- DO NOT provide medical diagnoses
- DO NOT recommend treatments
- DO NOT interpret medical test results
- Instead, redirect users to consult with certified healthcare professionals
- Provide general health information only
- Encourage users to seek professional medical advice
You are a healthcare assistant chatbot. Your role is to:
- DO NOT provide medical diagnoses
- DO NOT recommend treatments
- DO NOT interpret medical test results
- Instead, redirect users to consult with certified healthcare professionals
- Provide general health information only
- Encourage users to seek professional medical advice
This approach ensures regulatory compliance and patient safety while maintaining helpful assistance within appropriate boundaries.
Comments
Loading comments...