Read with the model in view
Use the visual as a map for the concepts, evidence, and decisions described in this note.
Students often treat AI mistakes as isolated surprises. A small error pattern gallery helps them see that many problems repeat: unsupported claims, hidden assumptions, shallow examples, outdated context, and answers that sound precise without enough evidence.
The gallery does not shame the tool or the student. It creates a shared learning record that makes review more concrete.
How to build the gallery
Ask each group to submit one short AI output that needed correction. For every entry, capture:
- Prompt context: What was the student trying to learn or produce?
- Observed issue: What part of the answer was weak, misleading, or incomplete?
- Evidence used: Which source, test, calculation, or expert judgement exposed the issue?
- Revision move: What changed in the improved answer?
Keep examples brief so the gallery remains easy to scan during future projects.
Teaching use
At the start of a new AI-supported assignment, let students review three gallery entries and predict which risks might appear in their own work. This turns responsible AI use into a practical habit rather than an abstract warning.
Learning outcome
Students learn to recognise recurring AI error patterns and document the evidence behind their corrections before relying on generated content.
Turn the note into a classroom decision
- Choose one concept from the note.
- Name an artifact that would make it observable.
- Write the next action and how you would verify it.