
Record a 6-minute walkthrough showing how you'd validate AI chatbot outputs against your actual pati
You'll have a video demonstration of your validation protocol—showing which outputs you'd trust, which you'd flag, and why—that you can share with stakeholders before rollout.
What you'll make
AI Validation Walkthrough Video
- Open: the validation question and why it matters
- Demo: testing 3-4 AI outputs in real time
- Analysis: which outputs passed, failed, and why
- Protocol: the repeatable steps you'd use next time
- Stakes: what this prevents before patient care
- Close: how stakeholders can use this approach
Upload a 6-minute screen recording showing you testing actual AI outputs with your validation criteria
The 4 parts
Each part ends with a check you can see.
Test AI Consistency on Real Queries
Run the same clinical questions multiple times to find inconsistency patterns that would concern stakeholders
Done when
You have tested 3-4 clinical questions 5 times each and documented every inconsistency with screenshots or copied text
Stress-Test AI with Your Trickiest Cases
Challenge AI with the clinical scenarios where you've seen colleagues make mistakes to find where it breaks down
Done when
You have 4-5 documented cases where AI gave concerning responses to your most challenging clinical scenarios
Check AI Claims Against Your Sources
Cross-reference AI's clinical recommendations against guidelines, literature, or protocols you actually use
Done when
You have verified 3-4 AI recommendations against your actual clinical sources and documented where AI was wrong, incomplete, or outdated
Record Your Validation Demo
Create a screen recording showing your validation protocol in action on real AI outputs
Done when
You have a 6-minute video showing validation of 3-4 AI outputs with clear trust/flag/reject decisions and rationale
Capstone: you put together your AI Validation Walkthrough Video.
Where this helps
An elective: a single lesson that sits outside rtwo's job certificates and doesn't count toward one.
About this lesson
What you'll be able to do
- Record a 6-minute video demonstrating your validation protocol on actual AI outputs
- Establish clear criteria for when to trust vs flag AI responses in your clinical context
- Document 3-5 systematic failure patterns you'd catch before stakeholder rollout
- Have a repeatable validation workflow you can teach to colleagues
Who it's for
- Healthcare professionals evaluating AI tools for clinical use
- Medical administrators preparing for AI pilot programs
- Clinical staff who need to validate AI recommendations before patient care
What it won't cover
- Building AI models or changing their training
- Regulatory compliance frameworks or legal liability
- Patient data privacy protocols
- Comparative analysis of different AI vendors
Skills you'll practise
- Output consistency testing
- Clinical contradiction detection
- Evidence-based validation
- Stakeholder communication