Course Challenge
Test your judgment on ten realistic classroom scenarios, then finalize a course AI policy and an AI-fluent unit plan you can use this term.
How the Challenge Works
- Knowledge check: ten classroom scenarios covering every module. Aim for at least 8 out of 10, and revisit any lesson where you missed a question.
- Capstone: finalize the course AI policy and unit plan you’ve been building, ready to use with real students.
Part 1: Knowledge Check
1. (Module 1) A colleague says, 'I'll teach my students AI by showing them every feature of our district's chatbot.' What's the stronger approach?
Features change quickly. The four competencies transfer to any tool.
2. (Module 1) Which task should a teacher keep entirely for themselves?
Sensitive, personal student writing calls for your own reading and response. The relationship is the point.
3. (Module 1) You want AI to suggest reading groups. What's the most diligent input?
De-identify, and use approved tools. AI rarely needs names to help.
4. (Module 2) AI generated a 20-question quiz with an answer key. What must happen before students take it?
Answer-key errors multiply across every student. Discernment comes before distribution.
5. (Module 2) What's the right order for AI-assisted feedback?
Judging first prevents your assessment from being anchored to AI's.
6. (Module 3) A student asks AI to write their reflection paragraph about what they learned in a lab. What's the delegation problem?
For learners, the work is often the point. Reflection is exactly the thinking the assignment is for.
7. (Module 3) Which activity best builds student discernment?
Error hunts build both discernment and subject knowledge.
8. (Module 4) Your syllabus says only 'Use AI responsibly.' What's the main problem?
Assignment-level permissions with reasons make expectations clear across students' different courses.
9. (Module 4) A detector flags a multilingual student's essay as 92% AI. What should you do?
Detectors have documented false-positive problems, including for non-native English writers. Decisions need observable evidence and a fair conversation.
10. (Module 5) What's the most powerful way to teach AI fluency throughout a unit?
Repeated practice and visible modeling build lasting habits.
Part 2: Capstone
Bring together your work from Modules 4 and 5 into two documents you can use this term.
- Course AI policy (from Lesson 4.1), finalized to include:
- Permission levels and where students find each assignment’s level
- Reasons tied to learning
- Disclosure expectations and an AI use statement template
- Privacy guidance and approved tools
- A fair, conversation-first integrity process (from Lesson 4.2)
- An invitation to ask questions
- AI-fluent unit plan (from Lesson 5.1), finalized to include:
- Your own AI-supported prep tasks, with the discernment checks you’ll run
- Every assignment’s purpose, AI level, and reason
- At least one description activity and one discernment activity for students
- Two modeling moments
- A note on how you’ll protect student data
- Get a second opinion. Share both with a colleague, coach, or (de-identified) an AI assistant acting as a skeptical department head. Revise.
Self-assessment checklist. You’re done when you can say yes to all of these:
- Every assignment in the unit has a purpose, an AI level, and a reason.
- No student data goes into a tool that isn’t approved for it.
- I read and judge student work before any AI is involved in feedback.
- My integrity approach doesn’t depend on a detector score.
- Students will see me use AI, including checking it and deciding not to use it.
- I’d be comfortable showing this plan to my students’ families.
Congratulations, You’ve Finished AI Fluency for Educators
You can now:
- Use AI fluently in your own planning, materials, and feedback, with student privacy protected
- Decide what to hand off and what to keep in teaching work
- Teach students to delegate, describe, discern, and practice diligence through short, subject-based activities
- Write a clear, assignment-level AI policy with reasons students understand
- Handle academic integrity fairly, without relying on detectors
- Model fluent AI use for your learners
Where to Go Next
- Redesign assessment: Teaching Above the Line goes deep on rewriting objectives and assessments for the age of AI.
- Strengthen your own discernment: Discernment: Judging AI Output covers verification, bias, and team review in depth.
- Review the foundations: AI Fluency: Foundations is a good course to recommend to colleagues just getting started.
The 4D AI Fluency Framework was created by Rick Dakan and Joseph Feller with Anthropic and is released under CC BY-NC-SA 4.0. See aifluencyframework.org.