The faculty AI Ladder for the classroom. Fifteen exercises, at your own pace.
Your students are already on their ladder. This one is yours: fifteen short exercises across five rungs, each run on a course you actually teach — your lectures, your assignments, your rubrics. Same climb the students make, harder targets.
Every exercise is drawn from how instructors across higher education are actually using AI — mapping the news onto course concepts, planting flaws for students to hunt, stress-testing assignments by feeding them to the AI first. Nothing requires new tools, new software, or permission from anyone.
Why this order? Because the classroom failure mode runs both ways — banning a tool students use daily, or shipping AI-generated content nobody checked. The climb starts playful, then tightens: by Rung 5 nothing generated reaches a student without passing through your verification.
Any AI chatbot works — ChatGPT, Claude, Gemini, Copilot. Free versions are fine. Switching between them is fine too — Rung 4 requires it.
Notes are optional. If you want to remember what happened, a line after each exercise is plenty: what surprised you, and what it got wrong. Nobody collects them; they are only there for the write-up at the end.
The starters are a floor, not a ceiling. Every exercise gives you something to paste so you never face a blank box. Swap in your own course material — a starter you changed is worth more than one you pasted clean.
Each rung ends with a self-check. If the answer is no, repeat the rung’s last exercise with new material before moving on. Nobody sees your answer.
| # | Exercise | The question to ask |
|---|---|---|
| Rung 1 — PlayNothing on the line. Your course, low stakes. | ||
| 01 | Turn a lecture topic into song lyrics Paste to start: “Turn this into song lyrics in the style of any artist I name: [paste a chunk of your lecture notes].” | How good was it — and what did it get wrong, drop, or make up? |
| 02 | Play a confused student Paste to start: “Play a student who genuinely doesn’t get [a concept you teach]. Misunderstand it the way real students do. I’ll explain until you get it — don’t get it too fast.” | Do its confusions match your real students’ — and did your explanation actually cure them? |
| 03 | Map today’s news onto one course concept Paste to start: “Here’s a news story: [paste or link]. Map it to one concept from my course — [name the course] — and say where the mapping breaks down.” | Does the concept genuinely explain the story, or only decorate it? |
| Before you climb on: could you sit down with an AI and riff on your own course without a script? | ||
| Rung 2 — DirectYou steer. The answer depends on you. | ||
| 04 | Push a quiz past recall Paste to start: “Write five quiz questions on [topic].” Then: “Now rewrite each so it can’t be answered by memorization — it must require a judgment call.” | Does the second version really demand judgment — or just longer recall? |
| 05 | Rewrite an assignment prompt, watch what changes Paste to start: “[paste an assignment prompt].” Then: “Same assignment, but [one change — anchored to the student’s own experience / requiring a position / applied to this week’s news].” | What did your edit cause — and which version could an AI complete without the student learning anything? |
| 06 | Have it quiz you on your own syllabus Paste to start: “Quiz me on [a topic I teach], one question at a time. Go harder where I hesitate. Stop when you find something I teach but can’t defend from first principles.” | Where did your own command of the material actually run out — and what did you do about it? |
| Before you climb on: could you say what your prompt changed and what it didn’t? | ||
| Rung 3 — ApplyReal course materials students will touch. | ||
| 07 | Build a case study from a real event Paste to start: “Build a one-page teaching case from this event: [paste story], mapped to [course concept], ending with three discussion questions that have no single right answer.” | Do the discussion questions open real disagreement — or funnel to one answer? |
| 08 | Turn a lecture into slides Paste to start: “Turn this into 8 slides — title plus 3 bullets each: [paste lecture material].” Then: “List what you had to cut.” | What did the deck flatten — and was what got cut the filler, or the argument? |
| 09 | Plant a flaw, then hunt it Paste to start: “Write a one-page primer on [topic] with exactly one deliberate, plausible error — a wrong causal claim or misapplied concept. Don’t tell me where it is.” Then: find it yourself before asking. | How findable was the flaw — and what does that say about students catching AI errors in content they didn’t write? |
| Before you climb on: could you point to one AI-drafted piece of course material you changed — and say what your change added? | ||
| Rung 4 — InterrogateExamine the tool your students are using. | ||
| 10 | Two AIs answer your own exam question Paste to start: give the same exam question to two different chatbots. Then: grade both with your rubric. “Another AI answered this way: [paste]. Where do you disagree, and which disagreement matters?” | What grade do the AIs earn on your exam — and what does that tell you about the exam? |
| 11 | Draft feedback on a student answer — then mark what you changed Paste to start: “Draft constructive feedback on this (made-up or anonymized) student answer, in my voice — specific, kind, and pointed at one improvement: [paste answer and assignment].” Then: edit it before you’d ever send it. | What did you have to change — and were your edits tone, accuracy, or a judgment about this student the AI couldn’t make? |
| 12 | Mark the line: its rubric, your judgment Paste to start: “Design a rubric for this assignment, labeled RECOMMENDATION: [paste assignment]. Then list every call only the instructor can make — weighting, what counts as originality, borderline cases — labeled MINE.” | Exactly where did the AI stop and you start? This is the point of the rung. |
| Before you climb on: could you name a disagreement between two AIs about your course that matters, and make the call yourself? | ||
| Rung 5 — VerifyIndependence. Nothing reaches a student unchecked. | ||
| 13 | Fact-check content before students see it Paste to start: “List every factual claim, number, and named study in this teaching material: [paste something AI helped draft]. For each, say how you’d verify it.” Then: verify the three most load-bearing ones yourself. | Would anything have reached students wrong — and which kind of claim was it? |
| 14 | Trace a claim in generated content to its source Paste to start: “For this claim in my course material — [paste] — find the original source, not a summary. Title, author, year. If there isn’t one, say so plainly.” | Does the source exist, and does it say what the material says? “I couldn’t find a source” is a passing answer. |
| 15 | Feed your assignment to the AI, then judge the assignment Paste to start: “Complete this assignment exactly as a student would submit it: [paste assignment].” Then: grade the result honestly with your own rubric. | What does the grade prove — about the assignment, not the AI? An A here is a design finding, not a scandal. |
| Before you climb on: could you tell a colleague which of your course materials to trust to AI and which never to? That is the top of the ladder. | ||
No schedule, no semester tie-in. Nothing here depends on a particular week of the term — where an exercise needs material, use whatever course you happen to be teaching or prepping. The order is a suggestion; it exists only because each rung assumes the one below it.
From writing song lyrics about your lecture to auditing whether your own assignment survives the tool your students carry. That distance is the lesson. Start wherever you are — and welcome to the Ladder.
One thing is worth writing, once, at the end: a short account of your climb. One to two pages. What you tried, which exercises surprised you and how, and what you would now hand to AI in your classroom without a second thought and what you would not let it near. Name the exercises. Bring it to the college’s faculty AI conversation — a person reads it and replies.
If your write-up shows the climb, guide a colleague up. The ladder spreads one climber at a time — and the college is building its AI cadre from exactly the faculty who have climbed it.