Frame the design decision
Use the sprint to reduce the biggest uncertainty—not to make a visually polished prototype. Decide the single question that must be answered by Friday.
Pathway 2 · Lesson 05120 min
AI Design Sprint Facilitation
Running 5-day sprints to discover and prototype AI use cases
Learning Objective
By the end of this lesson, you will be able to: Facilitate a condensed AI design sprint, apply key sprint methods (lightning demos, crazy 8s, storyboarding), and adapt the sprint format for remote and AI-assisted teams.
5
days from problem to tested prototype
Engage
"Most organisations spend months debating AI use cases. The AI Design Sprint compresses that to 5 days — with a tested prototype at the end. How does it work?"
Explore
Jake Knapp (inventor of the Design Sprint) explains the 5-day process — the foundation for the AI Design Sprint.
Explain
The 5-Day AI Design Sprint
Day 1 — Map: Understand the AI opportunity. Map the user journey. Identify the critical moment where AI could add the most value. Ask the long-term goal question: 'In 2 years, what does success look like?'
Day 2 — Sketch: Lightning demos (inspiring examples from other industries). Individual sketching. Crazy 8s (8 sketches in 8 minutes). Each person creates a detailed 3-panel storyboard.
Day 3 — Decide: Silent critique of all sketches. Dot voting. The Decider makes the final call. Create a single storyboard for the prototype.
Day 4 — Prototype: Build a realistic-looking prototype in one day. Fake it — use Figma, slides, or even paper. The goal is something testable, not something shippable.
Day 5 — Test: Interview 5 users. Watch them interact with the prototype. Look for patterns. Make the go/no-go decision.
✅
AI-Specific Sprint Adaptations
AI design sprints need three adaptations from the standard sprint: 1. Add an 'AI Audit' on Day 1 — what data exists? What's technically feasible? 2. On Day 3, add a 'Responsible AI' check — does the prototype raise fairness, transparency, or privacy concerns? 3. On Day 5, test for trust as well as usability — 'Do you trust this recommendation?' is as important as 'Can you use this feature?'
Elaborate · Worked Example
AI Sprint in Action: Customer Churn Prediction
Context
A SaaS company wants to use AI to predict and prevent customer churn.
The Situation
Day 1: Map reveals the critical moment is when a customer reduces usage — 30 days before they cancel. Day 2: Sketches explore 8 different interventions — from automated emails to AI-suggested human outreach. Day 3: Team decides to prototype 'AI-suggested personal outreach' — the account manager gets an AI-generated brief on why the customer is at risk and what to say. Day 4: Prototype built in Figma — a dashboard notification with AI-generated customer brief and suggested talking points. Day 5: 5 account managers test it. Key finding: They trust the AI brief but want to see the evidence behind it (explainability gap identified).
The Outcome
In 5 days: validated use case, tested prototype, clear design direction, and a specific improvement identified. Months of debate compressed into one week.
Try It · Reflect & Apply
Sprint planning: Choose an AI opportunity in your organisation. Write your Day 1 sprint map: What is the long-term goal? What is the critical moment in the user journey? What is the biggest risk to address? What question will your sprint answer?
Your notes are saved in your browser for this session.
Evaluate · Knowledge Check
1 / 3What is the primary output of Day 4 (Prototype) in a Design Sprint?
Design Practice Studio
AI Design Sprint Studio
Create the artefact
Prepare a sprint brief with long-term goal, target actor, critical workflow moment, key risk, responsible-AI checkpoint, prototype storyboard, and five-user test plan.
Test with evidence
During testing, capture evidence rather than explanations. What did users attempt first? Where did they hesitate? What did they think the AI knew or was allowed to do?
Spaced revisit
Hold a 30-minute evidence review within 48 hours. Decide whether to proceed, pivot, or pause—and document the specific evidence behind the choice.
External video references
Keep Learning on YouTube
Open in a new tab
Ready to move on?
Mark this lesson complete and finish the course.