Case study 03 · Teacher development
A 60-minute teacher workshop that moves the focus from making content with AI to judging it.
Teachers are being asked to use AI for lesson plans, messages and resources. But polished AI material can still have errors, the wrong level for the age group, hidden assumptions or privacy risks. Teachers need a simple routine for checking it, not just another list of prompts.
The workshop uses an AI-made lesson that is flawed on purpose (I wrote it myself). Teachers hunt for the problems, score it with a practical rubric, fix it, and explain why their version works better for students. Think of it as a spot-the-mistake game for grown-ups.
| Time | Activity | Evidence captured |
|---|---|---|
| 0–5 min | One-question baseline: Would you use this AI answer? | Anonymous initial judgement + reason |
| 5–15 min | Spot the hidden issues in a sample AI lesson | Individual annotation |
| 15–30 min | Evaluate with Accuracy–Age–Bias–Privacy–Pedagogy rubric | Completed evaluation rubric |
| 30–45 min | Redesign one activity and compare alternatives | Revised activity + rationale |
| 45–55 min | Peer review and improvement | Peer comment + revision note |
| 55–60 min | Exit challenge and action commitment | Post-response + next step |
