AI as Helper vs. AI as Crutch
The most important distinction in AI-era education.
The line between using AI as a learning tool and using it as a cognitive crutch is often invisible to students — and sometimes to teachers. But it is one of the most consequential distinctions in education right now. This lesson gives you the clarity to recognize it, talk about it, and teach it.
The Development Question
Every educational activity involves a tradeoff between the quality of the output produced and the development of the student doing the producing. A student who copies a perfect essay from the internet produces a better artifact than one who struggles through a rough draft — but the struggling student is developing skills the other is not. The artifact is not the point; the development is the point.
AI has made this tradeoff more acute because AI outputs are now so high quality that the production vs. development distinction matters more than ever. When AI writes a polished essay, solves a complex math problem, or generates a sophisticated analysis, the output may be excellent while the student's development is zero. The quality of the product is no longer a reliable signal of the quality of the learning.
This is the core challenge and the core teaching opportunity. Help students see that the real value of education is not in the products they produce but in the capabilities they build. AI can produce outputs for them indefinitely, but it cannot develop their judgment, their writing voice, their analytical reasoning, or their ability to navigate novel problems. Only they can do that — by doing the hard cognitive work themselves.
Ask students directly: "If you never had to write an essay yourself — because AI always did it for you — what do you think you'd be missing?" The answers are often surprisingly insightful. Students know, on some level, what they're trading away. Helping them articulate it is half the teaching.
The Spectrum of AI Use in Academic Work
AI use in academic work exists on a spectrum, not a binary. Understanding this spectrum helps you communicate more clearly with students about what is and isn't appropriate.
Clearly appropriate: using AI to understand a concept you're struggling with, so you can then do the work; asking AI to give feedback on a draft you've already written, then revising yourself; using AI to brainstorm options you then select from and develop; using AI to check grammar after you've written.
Requires judgment and context: using AI to help structure an essay outline (depends on whether structure-building is a learning objective); using AI-generated content as a starting point you then substantially rewrite; using AI for research when the assignment goal includes research skills.
Clearly problematic: submitting AI-generated text as your own writing; using AI to answer assessments designed to measure your understanding; letting AI do the cognitive work an assignment is designed to develop.
Your classroom policy should reflect this spectrum. Blanket bans don't prepare students for a world where AI is everywhere; blanket permission doesn't protect the developmental purposes of education. Thoughtful guidance about which uses serve learning and which bypass it is more valuable than any rule.
What Over-Reliance Looks Like
Over-reliance on AI for academic work has predictable signs that you can recognize. Students who have been heavily using AI for writing often struggle with in-class, unassisted tasks — the blank page problem becomes acute because they've stopped developing the muscle of starting from nothing. Their AI-assisted work may be polished but when asked to explain or defend it, they can't.
Over-reliance affects not just writing but any skill that AI can substitute for: mathematical reasoning, research navigation, critical analysis. The student who always asks AI for the answer stops developing the ability to reason through problems independently. This is not a character flaw; it is a predictable response to having an extraordinarily capable assistant available at all times.
Your job is not to punish this pattern but to illuminate it. Helping students see the connection between their AI habits and their independent capabilities — in concrete, non-judgmental terms — is what changes behavior. A student who realizes they can't explain their own essay, or solve a simpler version of a problem they "completed" with AI, has encountered the most powerful teacher available.
The Explain-Back Test
A simple classroom tool: after students complete any significant assignment, ask them to explain one key decision they made — about argument, structure, evidence, or solution strategy. Students who did the cognitive work can do this easily. Students who relied on AI often cannot. Frame this not as a gotcha but as a natural part of the learning process: "Help me understand your thinking here." The conversation is diagnostic and, often, educational for the student regardless of what it reveals.
The Math Mistake
Chapter 1 — Tool vs. Crutch Mentality
Margaret is struggling with her math homework. She copies all her word problems into an AI app that gives her answers. The next day in class, there's a quiz, and she can't solve anything on her own.
In the Level 2 version of this scenario, Margaret uses AI to break down complex word problems into smaller steps, then practices similar problems on her own — using AI as a tutor when she gets stuck, gradually decreasing her reliance as her competence increases. The contrast between these two approaches illustrates the helper vs. crutch distinction precisely.
- What is the difference between Margaret using AI to get answers vs. using AI to understand the process?
- How would you recognize Margaret's pattern in your own students? What would you see?
- What is a subject-area equivalent in what you teach — a skill students need to develop through practice that AI can now easily substitute for?
- How do you have a non-judgmental conversation with a student about over-reliance on AI?
The Helper vs. Crutch Checklist — Student Self-Assessment
Have students complete this self-assessment honestly and privately. Then discuss the results as a class — anonymously — using the aggregate patterns to open a conversation about what learning is actually for.
🔑 CCR for Your Classroom
Help students distinguish between AI helping them think and AI thinking for them. The first builds capability; the second borrows against it.
The most creative use of AI is as a genuine thinking partner — not a ghostwriter, but a sounding board that the student questions, challenges, and redirects.
Students who use AI as a crutch are making choices that affect their future selves. Help them see the long-term stakes, not just the short-term convenience.
Academic Integrity in the AI Era
Moving beyond "don't cheat" to building genuine understanding.
The conversation about academic integrity and AI is often framed as a detection problem. That's the wrong frame. The real question is not "can I catch it?" but "do my students understand why honesty in their work actually matters to them?" This lesson helps you have the conversation that changes minds, not just behavior.
Why Rules Aren't Enough
Academic integrity policies are necessary but insufficient. A student who follows AI rules only because they fear getting caught is not developing ethical judgment — they are developing compliance. Compliance breaks down when detection is unlikely; judgment holds even then.
The goal is to help students internalize the principles behind the rules. Why does honest work matter? Not because of the grade, not because of the policy, but because: the assignment is designed to develop capabilities the student will actually need; teachers need accurate information about what students know to give them appropriate support; submitting AI work as your own misrepresents your abilities in ways that may have real consequences (college applications, job interviews, professional situations where you are expected to do what your credentials say you can do).
This is not a lecture; it is a conversation. Students who reason through these principles themselves — through scenario discussion, case analysis, and genuine reflection — arrive at more durable understanding than students who are simply told the rules.
Transparency as a Classroom Norm
One of the most practical and powerful shifts you can make is normalizing transparency about AI use. When AI use is something students disclose on their work — "I used AI to brainstorm, then wrote this myself" or "I used AI for grammar checking but the content is mine" — it stops being a shameful secret and becomes a normal part of the learning process.
This norm has several benefits. It gives you actually useful information about your students' relationship with AI. It gives students a vocabulary for describing their process honestly. It reduces the temptation toward deception because disclosure is the expected and accepted default. And it prepares students for professional environments where AI disclosure is increasingly expected.
Starting this norm does not require elaborate policy. A simple note on your assignment prompt — "Please include a brief note on any AI tools you used and how" — is enough to establish the expectation. Review these notes not as surveillance but as information about your students' processes.
Kathi's own transparency about AI use in writing her book — described honestly in "The Elephant in the Room" — is a powerful example you can share with students. An accomplished author using AI tools and being fully transparent about it models exactly the norm you're trying to establish.
The Authentic Voice Question
One of the deepest questions AI raises about academic work is the question of authentic voice. Writing, in particular, is not just about producing text — it is about developing a voice, a way of thinking on the page, that is genuinely your own. AI can produce fluent, well-organized text, but it cannot produce your voice, because your voice emerges from your experiences, your thinking, and your long practice of putting ideas into words.
This matters because voice is not just an aesthetic quality; it is a marker of genuine intellectual engagement. A student whose essays always sound like polished AI output is not developing as a writer or thinker. Their work may score well by certain metrics while representing genuine atrophy in the capabilities that matter most.
Help students value their authentic voice by showing them the difference. Ask them to read a piece of their own unpolished writing and an AI-generated version on the same topic. What is present in theirs that is absent in the AI's? What specific experiences, observations, or opinions show up in their writing that the AI could never produce? This exercise often surprises students — they discover that their imperfect, unpolished writing contains something genuinely valuable.
The Transparency Note
Add one line to your assignment prompts: "Please briefly note any AI tools you used in completing this work and how you used them." Make clear that honest disclosure is not penalized — hiding it is. Over a few weeks, this simple practice normalizes transparency, gives you useful data, and opens natural conversations about process.
The Voice Recovery Exercise
Ask students to read a piece of AI-generated writing on a topic they know well, then identify everything that is missing — personal perspective, specific detail, genuine uncertainty, authentic voice. Then have them write one paragraph that could only have come from them. Often this simple exercise reconnects students with what makes their writing genuinely theirs.
The Perfect Paragraph
Chapter 2 — AI as a Helper: But You're Still the Author
Amy needs to write a paragraph about her summer vacation. She types "write a paragraph about summer vacation" into an AI tool and copies what it gives her word-for-word. It sounds really good, but it's not really her story.
The Level 2 version of this scenario involves a college application essay — an even higher-stakes example where a student submits AI-generated content as her personal statement, only to find that admissions officers are trained to identify inauthentic essays. The same principle applies: authentic voice is both more honest and ultimately more effective.
- What does authentic student voice look like in your subject area? How would you know it from an AI-generated equivalent?
- What would you say to Amy — not to shame her, but to help her understand what she's giving up?
- How might you design a writing assignment that makes authentic student voice essential rather than optional?
- What does the college application scenario in Level 2 suggest about the long-term consequences of AI substitution habits?
Talking Points for the AI Integrity Conversation
These are not scripts — adapt them to your voice and your relationship with your students. The most important thing is the authenticity of the conversation, not the specific words.
🔑 CCR for Your Classroom
The habit of honest self-assessment — "Is this really my thinking?" — is a critical skill that transfers far beyond AI use.
Students who develop authentic voices and genuine capabilities will be far more valuable and resilient in an AI-saturated world than those who can only manage AI outputs.
Model transparency consistently. If you use AI, say so. If you ask students to disclose, make disclosure safe and normal, not a confession.
Facilitating Rich Classroom Discussions About AI
The practical skills for leading conversations that actually change how students think.
The scenarios in Kathi Kersznowski's book The Educator's Guidebook for Teaching AI Literacy and Ethics are not meant to be worksheets. They are meant to spark genuine intellectual wrestling — the kind of messy, uncomfortable, productive conversation that changes how students think about AI, about honesty, about their own future. This lesson gives you the facilitation moves to make that happen.
Why Scenario-Based Discussion Works
AI ethics is not a topic where students benefit from being told the right answers. It is a domain of genuine complexity, competing values, and evolving norms where the skill to be developed is the ability to reason carefully through difficult situations — not the ability to recall a set of rules.
Scenario-based discussion works for AI ethics for several reasons. Concrete situations are more cognitively engaging than abstract principles. Characters in scenarios give students someone to identify with, argue against, and think through. Multiple perspectives naturally emerge when students try to justify their positions. And the messiness of real scenarios — where multiple values are in tension and there is no clear right answer — is itself educational: it prepares students for the real AI decisions they will face.
Kathi Kersznowski's book The Educator's Guidebook for Teaching AI Literacy and Ethics's 160 scenarios are specifically designed for this kind of discussion. They present situations with genuine ethical complexity, at two developmental levels, with discussion questions and talking points. The goal is always to provoke thinking, not to deliver conclusions.
Core Facilitation Moves
Effective discussion facilitation requires a different skill set from direct instruction. The goal is not to transfer your knowledge to students but to create the conditions in which students develop their own understanding through talk and argument.
Press for reasoning: when a student makes a claim, don't evaluate it — ask how they got there. "What makes you think that?" "What information are you drawing on?" "Is there another way to look at that?" Students who learn to articulate their reasoning develop more durable understanding than those who only state conclusions.
Introduce the complication: most good AI scenarios have a factor that complicates the initial intuition. If students quickly agree on an answer, surface the complication: "But what about X? Does that change anything?" "What if we also knew that..." This is not devil's advocacy for its own sake — it is teaching students that ethical situations are rarely simple.
Make space for uncertainty: some AI ethics questions don't have settled answers, and it is important for students to experience that. "I'm not sure about this one either — let's think through it together" models that complexity is real and intellectual humility is appropriate.
Connect to their lives: the most powerful discussions connect the scenario to students' actual experiences. "Has anyone been in a situation like this?" "What would you actually do?" "What would you tell a younger student about this?"
The silence after a hard question is not empty. It is thinking. Resist the urge to fill it. Students who are given time to think often say the most substantive things.
Using the Scenarios Flexibly
Kathi Kersznowski's book The Educator's Guidebook for Teaching AI Literacy and Ethics's scenarios are designed to be flexible. You don't have to use every question, run the full discussion, or follow the suggested structure. Some of the best classroom moments come from taking one question from a scenario and exploring it deeply rather than moving through all eight.
Scenarios work well as: opening hooks for a new unit (present the scenario before students have the conceptual framework — come back to it after); exit tickets (have students write one-paragraph responses to a scenario question as a check for understanding); Socratic seminars (use the Level 2 version as the text for a seminar discussion); debate setups (assign students positions and have them argue them from the scenario).
You also don't have to use every scenario in Kathi Kersznowski's book The Educator's Guidebook for Teaching AI Literacy and Ethics. Choose the ones that fit your students' current needs, your curriculum context, and the moment you're in. A scenario about AI bias hits differently after students have seen a real example of algorithmic bias in the news. A scenario about academic integrity has more traction at the beginning of a major assignment.
The Think-Pair-Share Upgrade
The classic think-pair-share works well for AI scenarios, but with one upgrade: after pairs share, ask each pair to report not just what they decided but where they disagreed. The disagreements are often more interesting than the agreements, and surfacing them makes the complexity of the scenario visible to the whole class.
Managing the Uncomfortable Moment
AI ethics discussions sometimes surface uncomfortable topics: cheating students or families know about, online privacy incidents, experiences with biased technology. Plan for this by establishing a classroom norm of "we discuss ideas, not people" and by being ready to redirect from specific accusations to the general principles at stake. "Let's talk about what the right thing to do would be in this situation" rather than "let's evaluate what [specific person] did."
The Group Project Dilemma
Chapter 1 — Collaboration, Transparency, and Fairness
Tricia is working on a group project about ecosystems with three classmates. Without discussing it with her team, she uses an AI tool to write most of the presentation slides the night before. The slides look polished, so the group decides to use them as-is. During the presentation, a classmate is asked a follow-up question about one of the slides — but they freeze, realizing they don't actually understand what was written.
This scenario is particularly rich for classroom discussion because it involves multiple ethical dimensions simultaneously: honesty with teammates, shared responsibility, understanding your own work, and fairness to other groups who worked without AI.
- How would you guide a discussion where students initially have strong, conflicting reactions to Tricia's choice?
- What facilitation move would you use if the class quickly agrees that Tricia was wrong and the discussion dies?
- How does this scenario connect to collaboration and fairness norms in your classroom more broadly?
- What is the most important thing you want students to take away from this discussion?
Your Discussion Facilitation Toolkit
Post these on the wall during discussion days. Over time, students internalize them and begin using them to push each other's thinking — which is the real goal.
🔑 CCR for Your Classroom
Discussion that presses for reasoning is critical thinking practice. Students who have to justify their positions develop more robust thinking than those who only state conclusions.
The best scenario discussions don't arrive at neat answers — they help students understand that complex situations require creative, contextual judgment.
Your facilitation choices model intellectual responsibility: taking ideas seriously, sitting with uncertainty, treating disagreement as productive rather than threatening.
Module 2 Knowledge Check
Consolidate your thinking before earning your certificate.
Module 2 Complete!
You've finished When AI Helps and When It Hurts. Your certificate and digital badge are ready below.
📬 Stay Connected
Want to hear about new courses, resources, and AI literacy updates from Kathi? Leave your email to join the community — completely optional, never spammy.
Course based on The Educator's Guidebook for Teaching AI Literacy and Ethics by Kathi Kersznowski
Share your professional development badge on LinkedIn or in your email signature.
Ready to continue your professional development?