Build an AI Health-Info Helper That Knows What to Refuse
Maps to: AI Application Builder · Physician · Public Health · Health Policy · Medical Researcher · Health Journalist
You're going to build an AI assistant that helps people understand health information from trusted sources like the CDC and NIH, and that knows what to refuse: it never diagnoses, never gives medical advice, and routes anything serious to a real clinician. The skill is safety-boundary design: deciding what a high-stakes tool must never do, then proving it holds on the dangerous cases. That's the scarcest thing AI application builders do as AI moves into health and other high-stakes fields, and doing one tells you fast whether building tools that fail safe is your kind of work.
How this shows up on a resume or college app
I built an AI assistant for [scope] that handles edge cases like emergency symptoms and requests for diagnosis by refusing and routing users to real clinicians, designing the safety eval and proving it fails safe. I learned that the hardest engineering problem in medical AI is what the tool should refuse to do.
When you finish, BuildMe drafts your Common App activity description from what you actually built.
The plan
- 1
Step 1
Pick a narrow scope + write the safety boundary
Pick a narrow health-info scope, then, first and before anything else, write the safety boundary: the disclaimer language, and the 'when to see a real clinician' and emergency-routing logic. This safety milestone is non-negotiable and comes first; it's the spine of the whole project.
- 2
Step 2
Build the system prompt + test 10 edge cases
Build the system prompt, referencing only authoritative sources (NIH, CDC, Mayo), and test it with 10 questions, including the dangerous edge cases on purpose: emergency symptoms, suicidality, requests for diagnosis. You're trying to break it.
- 3
Step 3
Design the refusal eval + iterate until it fails safe
The core judgment: design an eval set of dangerous edges, decide what the tool must REFUSE, and iterate the prompt until it routes correctly on EVERY one. Edge-case testing is humbling; it'll keep failing your safety checks. Every failure you catch is the whole point.
- 4
Steps 4–5
Publish, get 3 adults to review, document the limits
Publish it and share with 3 adults who can stress-test the edges and review it. Then write a one-page reflection on what you couldn't do safely; naming the limits is part of doing safety well.
Sign up and this plan gets personalized to your level, interests, and goal in about 15 seconds.