How educators can use artificial intelligence to design better lessons, guide meaningful practice, provide useful feedback, and help learners develop genuine understanding.
AI Can Generate a Lesson in Seconds. But Does That Mean Students Are Learning?
Artificial intelligence is changing education.
Today, an educator can ask AI to create a lesson plan, generate a quiz, explain a complicated concept, develop a worksheet, or produce a complete presentation.
What once required hours of preparation can sometimes be drafted in minutes.
But there’s an important question we need to ask:
Are we using AI to improve teaching, or simply to produce more educational content?
A beautifully formatted lesson doesn’t guarantee understanding.
A student who receives the correct answer hasn’t necessarily learned how to solve the problem.
An AI-generated explanation might be accurate, but that doesn’t mean every learner understands it.
And completing an AI-assisted assignment doesn’t automatically demonstrate independent capability.
This is the challenge I wanted to address when I developed Vibe Teaching, an educator-focused approach that grew out of my Vibe Learning framework.
The central idea is simple:
AI should help educators create better learning experiences while keeping human thinking, curiosity, judgment, and understanding at the center.
To make these ideas practical, I’ve created a free interactive resource:
Vibe Teaching — 50 Interactive Lessons & Teaching Guide
https://vibelearning.ca/vibe-teaching.php
The guide provides a structured approach to designing AI-supported learning experiences, making classroom decisions, developing meaningful activities, and reflecting on evidence of learning.
But before exploring the resource, let’s examine how Vibe Teaching works and why it matters.
1. What Is Vibe Teaching?
Vibe Teaching is an approach to using AI to design and guide active learning experiences.
The educator defines the learning objectives, designs meaningful activities, determines how AI can help, and evaluates evidence of student understanding.
AI can assist with explanations, examples, questions, feedback, differentiation, and lesson preparation.
But the learner remains responsible for thinking, making decisions, attempting tasks, explaining reasoning, and demonstrating what they’ve learned.
The three essential roles
The educator: Designs the learning experience, sets expectations, evaluates progress, and decides when intervention is appropriate.
The learner: Explores ideas, attempts problems, makes choices, asks questions, reflects, and demonstrates understanding.
AI: Supports the process through guided explanations, practice opportunities, feedback suggestions, and alternative approaches.
The key is maintaining the distinction between assistance and learning.
AI can support the learning process. It should not automatically replace the thinking the learner needs to develop.
This applies to traditional classrooms, online courses, workplace training, tutoring, professional development, and self-directed education.
2. From Vibe Learning to Vibe Teaching
In 2025, as AI-assisted development and vibe coding became increasingly prominent, I began exploring how conversational AI could support a more structured approach to learning.
I developed the Vibe Learning framework around the idea that AI should encourage active exploration and understanding rather than simply deliver answers.
Shortly afterward, I extended those ideas into Vibe Teaching, the educator-facing counterpart.
The two approaches share the same philosophy but address different responsibilities.
Vibe Learning: The learner’s perspective
The central question is:
How can I use AI to learn more effectively?
Learners set goals, ask questions, explore concepts, practise skills, receive feedback, and check whether they can apply their knowledge independently.
Vibe Teaching: The educator’s perspective
The central question becomes:
How can I use AI to help someone else learn more effectively?
Educators design questions, challenges, feedback, explanations, and assessments that help learners develop capability.
| Vibe Learning | Vibe Teaching |
|---|---|
| Learner directs personal exploration | Educator designs the learning environment |
| AI supports individual understanding | AI supports instructional decisions |
| Learner practises and reflects | Educator creates opportunities to practise and reflect |
| Learner checks personal mastery | Educator gathers evidence of understanding |
| Focus on learning effectively | Focus on teaching effectively |
These are complementary perspectives.
Both emphasize a principle that I believe is increasingly important:
The purpose of educational AI is not to remove human effort. It’s to make that effort more meaningful and productive.
3. The Difference Between AI-Generated Content and AI-Supported Learning
Consider two approaches to teaching JavaScript conditionals.
Approach A: Content generation
An educator asks AI:
“Create a 30-minute lesson about JavaScript if statements.”
The AI generates:
- A definition.
- Several examples.
- A presentation outline.
- Five quiz questions.
- A summary.
The material might be useful.
But students could still leave without understanding how conditional logic works.
Approach B: Learning experience design
Instead, the educator asks:
“Design a learning activity where beginners predict how JavaScript conditional statements behave, test their predictions, explain unexpected results, and apply the concept to a new problem.”
Now the focus changes.
Rather than simply presenting information, the lesson requires students to do something with it.
A practical example
Show students:
const score = 75;
if (score >= 80) {
console.log("Excellent");
} else if (score >= 60) {
console.log("Passed");
} else {
console.log("Try again");
}Before running the code, ask:
What will this program display, and why?
Have learners make a prediction.
Then execute the code.
Ask them to change score to 80.
What happens?
What about 59?
What about the string "75"?
Now students are exploring behavior, identifying assumptions, and developing reasoning skills.
The AI can support the activity by generating additional examples or suggesting useful questions.
But the learning comes from the student’s engagement with the problem.
4. The Vibe Teaching Learning Loop
Vibe Teaching uses a flexible eight-stage instructional loop.
<box border radius=”xl” padding={3} gap={2}>
<title size=”lg”>The Vibe Teaching Learning Loop</title>
<grid columns={2} gap={2}>
{#each [{n:”01″,t:”Orient”,d:”Define the outcome and evidence”},{n:”02″,t:”Understand”,d:”Connect ideas to prior knowledge”},{n:”03″,t:”Predict”,d:”Commit to an idea before revealing results”},{n:”04″,t:”Apply”,d:”Attempt a meaningful task”},{n:”05″,t:”Feedback”,d:”Identify a specific next improvement”},{n:”06″,t:”Reflect”,d:”Explain decisions and discoveries”},{n:”07″,t:”Transfer”,d:”Apply learning in a new situation”},{n:”08″,t:”Refine”,d:”Adjust the teaching based on evidence”}] as item}
<grid-item>
<box background=”surface-secondary” padding={3} radius=”lg” gap={1}>
<text color=”secondary” size=”xs”>{item.n}</text>
**{item.t}**
<text size=”xs” color=”secondary”>{item.d}</text>
</box>
</grid-item>
{/each}
</grid>
</box>
These stages are not intended to become a rigid checklist.
Educators can return to earlier steps when learner responses reveal gaps or misconceptions.
Let’s examine how the stages work in practice.
Orient: Define what success looks like
Before designing an activity, ask:
What should learners be able to do by the end?
What evidence would demonstrate that capability?
For example:
“Students will be able to write a JavaScript conditional that categorizes a numerical score and explain why each branch executes.”
This is more useful than simply stating:
“Students will understand conditionals.”
The first outcome identifies an observable action.
Understand: Connect with prior knowledge
Students need a starting point.
Use familiar examples and diagnostic questions.
For conditional logic, you might begin with everyday decisions:
“If the temperature is below freezing, wear a winter coat.”
Then connect the decision structure to JavaScript.
Predict: Encourage thinking before revealing answers
Prediction creates an opportunity for learners to commit to an explanation.
Ask what they expect before showing the result.
This helps reveal misconceptions.
Apply: Practise the concept
Students should perform a task, not just observe a demonstration.
For example:
“Write a conditional that identifies whether a customer qualifies for a discount.”
Feedback: Identify the next improvement
Feedback should help the learner understand what to change.
Instead of:
“Incorrect.”
Try:
“Your condition checks the discount amount, but the requirement is based on the purchase total. Which variable should determine eligibility?”
Reflect: Explain the reasoning
Ask learners:
What did you predict?
What happened?
Why was your original reasoning correct or incorrect?
What would you do differently?
Transfer: Use the concept somewhere new
A learner who can repeat an example may still struggle with a different problem.
Try a new scenario.
For example:
“Use conditional logic to determine shipping costs based on order value.”
Refine: Improve the next learning experience
The educator examines the evidence.
Which concepts were understood?
Where did learners struggle?
What support was useful?
What should change next time?
This closes the loop between teaching decisions and observable learning.
5. Use AI to Ask Better Questions
One of the most valuable applications of AI in teaching is generating questions that reveal student understanding.
Consider this prompt:
“Explain JavaScript arrays.”
AI can produce an explanation.
But an educator might get more instructional value from asking:
“Generate five questions that reveal whether students understand the difference between an array index and an array value.”
The second request focuses on diagnosing understanding.
A reusable teaching prompt
ROLE:
Act as an experienced instructional coach.
TOPIC:
[Insert topic]
AUDIENCE:
[Describe learners and prior knowledge]
LEARNING OUTCOME:
[What learners should be able to do]
TASK:
Create five questions that reveal understanding.
REQUIREMENTS:
- Begin with a simple prediction.
- Include one common misconception.
- Ask learners to explain their reasoning.
- Include one practical application.
- End with a new transfer problem.
IMPORTANT:
Do not immediately reveal the answers.
Provide an educator answer guide separately.This prompt can be adapted to programming, mathematics, science, business, communication, or other subjects.
The educator should review the questions for accuracy, appropriateness, and alignment with the learning outcome.
6. Create Hint Ladders Instead of Giving Away Answers
A common challenge in AI-assisted education is that AI can provide complete solutions too quickly.
Sometimes students need a hint, not the answer.
A hint ladder provides progressively more specific support.
For example, imagine a student is struggling with this JavaScript problem:
“Write a function that returns the largest number in an array.”
Level 1: Guiding question
“What information do you need to keep track of as you examine each number?”
Level 2: Conceptual hint
“Think about storing the largest value you’ve encountered so far.”
Level 3: Partial implementation
function findLargest(numbers) {
let largest = numbers[0];
// Compare the remaining numbers.
return largest;
}Level 4: Worked explanation
If the learner still needs support, demonstrate a complete solution and explain the reasoning.
Afterward, provide a new problem to check understanding.
For example:
“Now write a function that returns the smallest number.”
The goal is not to withhold help.
It’s to provide support that matches the learner’s current needs while preserving opportunities for independent thinking.
AI prompt for generating hint ladders
Create a four-level hint ladder for this problem:
[INSERT PROBLEM]
Level 1:
A guiding question.
Level 2:
A conceptual hint.
Level 3:
A partial example.
Level 4:
A complete explanation.
Each level should provide more support.
Also create a different follow-up problem
that checks whether the learner can
apply the concept independently.7. Make Feedback Specific and Actionable
Feedback is one of the areas where AI can assist educators.
But useful feedback must do more than identify mistakes.
It should explain what matters and suggest a manageable next action.
Consider a student’s written response:
“JavaScript arrays store only numbers.”
An unhelpful response might be:
“Wrong. Arrays can store different types.”
A more useful response could be:
“You correctly identified that arrays store collections of values. However, arrays are not limited to numbers. Can you create an array containing a string, a number, and a Boolean value?”
This feedback:
- Acknowledges what the student understood.
- Identifies the specific misconception.
- Provides a practical next step.
- Encourages the learner to demonstrate understanding.
A feedback template
What worked: Identify something accurate or effective.
What needs attention: Name the specific gap.
Next action: Suggest one manageable improvement.
Follow-up check: Ask the learner to apply the idea independently.
AI can help draft feedback using this structure, but educators should review it before sharing.
Avoid entering identifiable student work or sensitive information into AI services without appropriate authorization and privacy safeguards.
8. Design Assessments That Measure Learning, Not AI Output
AI-generated assignments create an important assessment challenge.
A student may submit an impressive result without being able to explain how it was produced.
This doesn’t automatically mean the student learned nothing.
But it means the final product alone may not provide enough evidence.
Example: A JavaScript coding assignment
Traditional assessment:
“Build a working calculator.”
The educator evaluates the finished application.
A Vibe Teaching approach might include additional evidence:
Planning: Explain the calculator’s required behavior.
Prediction: Describe what should happen for different inputs.
Implementation: Build the calculator.
Testing: Demonstrate several test cases.
Explanation: Describe how the main functions work.
Transfer: Modify the application to support a new operation.
This approach helps distinguish between producing an artifact and demonstrating understanding.
A compact assessment rubric
| Criterion | Evidence |
|---|---|
| Correctness | Application behaves as expected |
| Reasoning | Learner explains key decisions |
| Testing | Learner checks normal and edge cases |
| Revision | Learner improves a specific weakness |
| Transfer | Learner applies the skill to a new task |
The exact weighting should depend on the learning outcome.
The important point is that assessment should measure the capability the course intends to develop.
9. A Complete 45-Minute Vibe Teaching Lesson
Let’s create a practical example educators can adapt.
Topic: JavaScript Conditional Logic
Audience: Beginner programming students
Learning outcome: Students can write and explain a conditional statement that handles multiple outcomes.
| Time | Activity | Purpose |
|---|---|---|
| 0–5 min | Prediction challenge | Reveal prior knowledge |
| 5–12 min | Guided explanation | Introduce conditional logic |
| 12–22 min | Individual coding task | Practise independently |
| 22–30 min | Feedback and revision | Address misconceptions |
| 30–40 min | New scenario challenge | Test transfer |
| 40–45 min | Exit ticket | Gather evidence |
Opening prediction
Show students:
const age = 18;
if (age > 18) {
console.log("Adult");
} else {
console.log("Other");
}Ask:
“What will the program display?”
Some learners may overlook the difference between > and >=.
This creates an opportunity to discuss comparison operators.
Guided practice
Ask learners to modify the program so that age 18 is included in the first branch.
Independent task
Create a ticket pricing function with three categories:
- Under 13: Child ticket.
- Ages 13–64: Standard ticket.
- Ages 65 and above: Senior ticket.
Feedback
Have students test boundary values such as 12, 13, 64, and 65.
AI can help suggest test cases, but learners should predict results and inspect the code themselves.
Transfer challenge
Ask students to create a new conditional for shipping fees based on purchase totals.
Exit ticket
“Explain why testing boundary values is important when writing conditional statements.”
This final question provides evidence of conceptual understanding beyond whether the original code ran successfully.
10. Five Practical Ways Educators Can Use AI Today
1. Generate alternative explanations
Ask AI to explain a concept using different examples or levels of complexity.
Review the explanations for accuracy and suitability.
2. Create practice variations
Generate additional exercises that target the same skill with different contexts.
For example, teach percentages using discounts, attendance rates, and survey responses.
3. Develop formative assessments
Ask AI to suggest questions that expose common misconceptions.
Use learner responses to guide the next instructional decision.
4. Prepare differentiated support
Create guiding questions, partial examples, and more detailed explanations for learners who need different levels of assistance.
Maintain comparable learning goals where appropriate.
5. Reflect on lesson effectiveness
Use anonymized classroom observations to brainstorm improvements.
For example:
“Students could identify a function but struggled to explain its return value. Suggest three short activities that reveal this distinction.”
AI can help generate options.
The educator decides which options are appropriate.
11. Introducing the 50-Lesson Vibe Teaching Guide
To help educators move from ideas to practice, I developed a free interactive teaching guide with 50 lessons organized into 10 modules.
Each lesson follows a practical structure:
Explore a teaching principle.
Understand the idea and why it matters.
Make a classroom decision.
Consider a realistic teaching situation and choose an instructional response.
Design an activity.
Adapt the principle to your own subject and learners.
Transfer and reflect.
Consider how the approach could work in another context and identify what evidence you would examine.
The guide is designed to support actual teaching decisions rather than simply provide information about AI.
What you can do inside the guide
- Explore 50 interactive lessons.
- Work through realistic classroom situations.
- Design activities for your own learners.
- Record decisions and reflections.
- Search and filter lessons.
- Bookmark relevant topics.
- Track your progress.
- Export a workbook.
- Save and restore progress backups.
- Access a printable guide.
Your responses and progress are stored in your browser, so exporting a backup is useful if you want to preserve your work.
The guide is a professional learning resource, not an accredited teaching qualification.
Explore the complete resource:
https://vibelearning.ca/vibe-teaching.php
12. What Will You Learn Across the 10 Modules?
The guide progresses through important areas of AI-supported instructional design.
Module 1: Purpose and Learner Agency
Learn to keep learner action at the center of teaching.
Explore how to define AI’s role, encourage questions, make uncertainty discussable, and preserve meaningful learner choice.
Module 2: Outcomes and Lesson Design
Create observable learning outcomes, baseline activities, aligned assessments, and realistic lesson plans.
Module 3: AI Coaching and Prompts
Develop coaching prompts, use focused questions, create hint ladders, set response constraints, and recognize when AI provides too much assistance.
Modules 4–5: Explanation, Practice, and Learning Evidence
Explore how to explain difficult ideas, build useful practice sequences, support collaboration, and collect evidence of understanding.
Module 6: Feedback and Assessment
Design feedback that leads to revision, distinguish supported performance from independent capability, and develop practical rubrics and exit tasks.
Modules 7–8: Responsible and Inclusive AI Use
Consider accuracy, privacy, bias, accessibility, different learning needs, and appropriate alternatives when AI access is limited.
Module 9: Projects and Transfer
Design projects with clear outcomes, meaningful checkpoints, and opportunities for learners to apply knowledge in new situations.
Module 10: Reflection and Course Improvement
Use evidence from teaching practice to refine activities, improve feedback, and develop more effective learning experiences.
Each module encourages educators to move beyond reading and create something they can use.
13. Design Your Own AI-Supported Lesson
The Vibe Teaching resource also includes a Lesson Design Studio.
This provides a practical way to turn the framework into an instructional activity.
A useful lesson design should answer several questions.
What will learners be able to do?
Define a measurable outcome.
What do learners already know?
Identify prerequisites and likely misconceptions.
What will learners attempt?
Design a meaningful activity.
How will AI help?
Choose a specific supporting role.
What evidence will demonstrate understanding?
Plan an observable check.
What happens when learners struggle?
Prepare feedback and graduated support.
How will you check transfer?
Include a new situation where learners apply the skill.
Reusable lesson design prompt
You are assisting an educator with lesson design.
SUBJECT:
[Insert subject]
TOPIC:
[Insert topic]
LEARNER LEVEL:
[Describe experience]
DURATION:
[Available time]
LEARNING OUTCOME:
[Observable skill]
DESIGN A LESSON THAT INCLUDES:
1. A short diagnostic activity.
2. One clear explanation.
3. A prediction question.
4. A practical learner task.
5. A four-level hint ladder.
6. A feedback and revision opportunity.
7. An independent transfer challenge.
8. An exit ticket.
REQUIREMENTS:
- Preserve learner decision-making.
- Avoid revealing answers too early.
- Include observable evidence of learning.
- Identify likely misconceptions.
- Suggest how the educator can respond.
OUTPUT:
A timed lesson plan with activities,
educator guidance and assessment criteria.This is a starting point, not a substitute for instructional judgment.
Educators should adapt the resulting plan to their learners, curriculum, institutional requirements, and available resources.
14. Responsible AI Use Is Part of Good Teaching
AI can support education, but it also introduces challenges.
Educators need to consider accuracy, fairness, privacy, accessibility, and appropriate levels of assistance.
Accuracy
AI-generated explanations can contain mistakes.
Review factual claims, calculations, examples, and references before using them.
Privacy
Avoid sharing identifiable student information with AI services unless the use is authorized and complies with applicable policies.
Bias
AI-generated examples may contain assumptions that don’t reflect the diversity of learners.
Review examples for missing perspectives and unnecessary stereotypes.
Accessibility
Not every learner has the same technology access, reading needs, or learning preferences.
Provide suitable alternatives and accessible materials.
Academic integrity
Make expectations clear.
Students should understand which kinds of AI assistance are permitted and what work they must complete independently.
Human oversight
The educator remains responsible for instructional decisions and assessment.
Responsible AI-supported teaching isn’t simply about using AI safely. It’s about ensuring the technology serves meaningful educational goals.
15. Start Small: A Seven-Day Vibe Teaching Challenge
You don’t need to redesign an entire course to begin.
Start with one lesson.
Day 1: Choose a Learning Outcome
Select a topic and identify one skill learners should demonstrate.
Day 2: Create a Prediction Activity
Develop a question that asks learners to commit to an answer before seeing the solution.
Day 3: Build a Hint Ladder
Create several levels of support for a challenging task.
Day 4: Improve Feedback
Rewrite generic feedback so it identifies a specific gap and a next action.
Day 5: Design a Transfer Challenge
Create a new situation that requires learners to apply the same underlying concept.
Day 6: Teach or Pilot the Activity
Try the activity with an appropriate group or review it with a colleague.
Collect observations about learner attempts and responses.
Day 7: Reflect and Improve
Ask:
What evidence of learning did I observe?
Where did learners struggle?
Did AI assistance support or replace thinking?
What should I change next time?
The goal is to build a repeatable teaching practice through small, evidence-informed improvements.
16. Why Vibe Teaching Matters for the Future of Education
As AI becomes more capable, generating educational content will become increasingly easy.
But creating content and creating learning are not the same thing.
The challenge for educators isn’t simply learning to use AI tools.
It’s learning to design experiences where those tools contribute to deeper understanding.
We should be asking:
Can learners explain what they know?
Can they recognize mistakes?
Can they evaluate information?
Can they solve unfamiliar problems?
Can they apply knowledge without relying on a generated answer?
Can they reflect on their decisions?
These questions matter far more than how quickly a lesson plan can be generated.
I believe AI offers educators an opportunity to spend less time producing repetitive materials and more time focusing on instructional decisions, feedback, meaningful practice, and learner development.
But realizing that opportunity requires deliberate design.
That’s the purpose of Vibe Teaching.
17. Start Exploring Vibe Teaching Today
If you’re an educator, trainer, tutor, course creator, or someone interested in AI-supported education, I encourage you to explore the Vibe Teaching framework.
You don’t need to adopt every technique immediately.
Start with one lesson.
Try one prediction activity.
Create one better feedback prompt.
Design one independent transfer challenge.
Observe what learners actually do.
Then refine your approach.
The 50-Lesson Vibe Teaching Guide provides a structured way to develop these practices through interactive activities, classroom decisions, lesson design, and reflection.
Explore Vibe Teaching:
https://vibelearning.ca/vibe-teaching.php
Explore the Vibe Learning Framework:
Discover Free Interactive AI and Coding Learning Tools:
https://discoveryvip.com/guides.php
The future of teaching with AI shouldn’t be about replacing educators or eliminating learner effort. It should be about creating better opportunities for people to think, explore, practise, and understand.
About the Author
Laurence Svekis is an educator, web developer, bestselling author, Google Developer Expert specializing in Google Workspace, and Google Champion.
With over 20 years of experience in web development and online education, Laurence has taught more than 3 million students worldwide.
He specializes in JavaScript, Google Apps Script, artificial intelligence, automation, instructional technology, and practical online learning.
Laurence developed the Vibe Learning framework and its educator-focused counterpart, Vibe Teaching, to explore how AI can support active learning, guided instruction, meaningful feedback, and independent understanding.
Through VibeLearning.ca and DiscoveryVIP, he creates interactive learning tools and educational resources for students, educators, and lifelong learners.
Join the Conversation
Are we using AI to help students learn—or simply to help them complete assignments faster?
AI can generate explanations, quizzes, presentations, and entire lessons.
But genuine learning still requires curiosity, practice, reasoning, feedback, and reflection.
I believe the most valuable role of AI in education is to help educators create more opportunities for these experiences.
What do you think?
How should educators balance AI assistance with independent student thinking?
Share your experiences and ideas.
Teach with AI. Make room for thinking.
#VibeTeaching #VibeLearning #ArtificialIntelligence #AIInEducation #EdTech #TeachingWithAI #GenerativeAI #InstructionalDesign #EducationInnovation #ActiveLearning #PersonalizedLearning #TeacherDevelopment #AIForEducators #LaurenceSvekis