.png)
You know the reality: thirty students, thirty different reading levels, thirty different attention spans, and one lesson plan. Differentiated instruction has always been the answer to reaching every learner, but finding the time to create multiple versions of the same material? That has always been the problem. AI is changing that equation fast. According to a 2025 EdWeek Research Center survey, 50% of teachers have now received at least one professional development session on AI, nearly double the rate from early 2024. Teachers are not just curious about AI anymore — they are using it to solve real classroom problems, and differentiation is at the top of the list.
This article walks you through ten practical, classroom-tested examples of differentiated instruction with AI — organized by readiness, interest, and learning profile. Each example includes what the differentiation looks like in practice, the AI tools that make it work, and prompts you can adapt today.
What is differentiated instruction, and why does AI make it easier?
Differentiated instruction is the practice of adjusting content, process, or product to match students' readiness levels, interests, and learning profiles. Developed by Carol Ann Tomlinson, this framework asks teachers to meet students where they are rather than delivering one-size-fits-all lessons. The three main ways to differentiate are by readiness (how prepared a student is for the content), interest (what motivates a student), and learning profile (how a student learns best).
The challenge has never been the concept — it is the workload. Creating three versions of a worksheet, adjusting reading levels for a passage, and designing choice boards for twenty-five students takes hours most teachers simply do not have. AI tools now handle the production side of differentiation in minutes, freeing teachers to focus on the decisions that matter: who gets which resource, when to adjust, and what the underlying learning gap actually is.
Platforms like TeacherPlug, an AI learning platform for teachers, take this even further by providing structured tutorials on how to use AI tools for differentiation, curated prompt libraries, and step-by-step guides tailored to real classroom scenarios. Instead of figuring out AI on your own, TeacherPlug walks you through the process so you can start differentiating faster and with more confidence.
Examples of differentiated instruction by readiness
Differentiating by readiness means adjusting the complexity, scaffolding, or depth of a task based on what students already know and can do. AI excels here because it can rapidly generate tiered versions of the same material.
1. Tiered reading passages for a science lesson
The scenario: A 5th-grade teacher is preparing a lesson on the water cycle. Some students read at a 3rd-grade level, most are on grade level, and a few are ready for more advanced content.
The AI differentiation: Using ChatGPT or Claude, the teacher pastes the original grade-level passage and prompts the AI to create three versions:
Below grade level: Shorter sentences, a built-in word bank (evaporation, condensation, precipitation), and comprehension questions focused on recall
On grade level: The original passage with open-ended questions asking students to explain each stage in their own words
Above grade level: An extended passage that introduces the concept of transpiration and asks students to compare the water cycle on Earth with what scientists hypothesize about water cycles on other planets
Why it works: All three groups learn the same core concept — the water cycle — but the cognitive demand and reading complexity match each group's readiness. The teacher spends five minutes prompting the AI instead of forty-five minutes rewriting passages manually.
2. Scaffolded math problem sets
The scenario: A 7th-grade math class is working on solving multi-step equations. Some students are still shaky on one-step equations, while others are ready for equations with variables on both sides.
The AI differentiation: The teacher uses an AI tool to generate three problem sets from the same learning objective:
Support tier: Part-completed equations with worked examples beside each problem, plus a reference card showing the order of operations
Core tier: Standard multi-step equations with no scaffolding
Extension tier: Multi-step equations embedded in word problems that require students to set up the equation themselves before solving
Why it works: The scaffolded tier removes barriers to entry without reducing the mathematical thinking required. Students still solve equations — they just get more support getting started. Tools like Diffit and MagicSchool AI can generate these tiers automatically from a single topic name, and TeacherPlug's prompt library includes ready-made prompts for creating tiered math activities across grade levels.
3. Differentiated exit tickets for formative assessment
The scenario: After a high school history lesson on the causes of World War I, the teacher wants to check understanding across all ability levels.
The AI differentiation: The teacher prompts AI to create three exit ticket versions from the same content:
Recall level: "List three causes of World War I."
Explanation level: "Explain how the alliance system contributed to the outbreak of World War I."
Evaluation level: "Which cause of World War I do you think was most significant? Support your argument with at least two pieces of evidence."
Why it works: Every student demonstrates understanding of the same topic, but the depth of thinking scales to match readiness. The teacher gets diagnostic data on every student, not just the ones who raise their hands. This approach aligns directly with Bloom's Taxonomy, moving from knowledge and comprehension at the support level to analysis and evaluation at the extension level.
Examples of differentiated instruction by interest
Differentiating by interest means giving students choices that connect learning to what they care about. AI makes this practical by generating multiple options quickly.
4. AI-generated choice boards for a writing unit
The scenario: A middle school English teacher is assigning a persuasive writing piece. Instead of one prompt for everyone, the teacher wants students to write about something they genuinely care about.
The AI differentiation: The teacher prompts ChatGPT: "Create a choice board with nine persuasive writing prompts for 8th graders. Include topics related to sports, technology, the environment, social media, animal welfare, school policies, music, gaming, and food. Each prompt should require students to take a clear position and support it with at least three reasons."
The AI generates a 3×3 choice board in under a minute. Students pick the topic that interests them most, and the teacher can add constraints (word count, required sources) as needed.
Why it works: The learning objective — writing a persuasive essay — stays the same for every student. But because students choose a topic they care about, engagement and effort increase. Research consistently shows that student choice improves motivation and writing quality. AI simply removes the bottleneck of creating nine different prompts manually.
5. Interest-based research projects with AI-curated resources
The scenario: A 4th-grade class is studying animal habitats. The teacher wants each student to research a habitat that fascinates them.
The AI differentiation: The teacher uses AI to generate a starter resource pack for each habitat option — desert, ocean, rainforest, arctic, and grassland. Each pack includes:
Three kid-friendly research questions
A vocabulary list with simple definitions
Suggested search terms for finding reliable sources
A graphic organizer template for note-taking
Students choose their habitat and receive a resource pack tailored to that topic.
Why it works: Rather than spending an entire class period helping students find starting points, the teacher provides structured entry points generated in minutes. Students with strong research skills can go beyond the provided questions, while students who need more structure have a clear path forward. This is the kind of differentiated instruction strategy that TeacherPlug's material generators are designed to support — creating curriculum-aligned resources quickly so teachers can focus on guiding the learning.
6. Connecting math to student interests
The scenario: A high school algebra teacher notices that some students disengage during abstract equation practice. The teacher wants to make the same skill set relevant to different student interests.
The AI differentiation: The teacher prompts AI to rewrite a set of linear equation word problems in four interest-based contexts:
Sports: Calculating a basketball player's points-per-game average and predicting season totals
Music: Determining how many songs fit on a playlist given different song lengths and a time limit
Business: Figuring out break-even points for a student-run bake sale
Travel: Calculating travel time and fuel costs for a road trip
Each version practices the same algebraic skill — setting up and solving linear equations — but the context changes to match what different students find engaging.
Why it works: Students who struggle with abstract math often succeed when the same problem is embedded in a context they understand and care about. AI generates these contextual variations in seconds, a task that would take a teacher significant time to do for each unit.
Examples of differentiated instruction by learning profile
Differentiating by learning profile means adjusting how students engage with content based on how they learn best. This includes considering factors like learning preferences, processing speed, and specific learning needs.
7. Multiple means of representation with UDL and AI
The scenario: A 6th-grade teacher introducing the concept of photosynthesis wants to present the content in multiple formats so every student has an accessible entry point.
The AI differentiation: Following the Universal Design for Learning (UDL) framework, the teacher uses AI to create:
A step-by-step text explanation with bolded key terms and a built-in glossary
A labeled diagram prompt that the teacher uses to create or find a matching visual
A sequencing activity where students arrange the steps of photosynthesis in order using cut-and-paste cards
A discussion guide with partner talk prompts for students who process information best through conversation
All four resources cover the same content. Students can use one or rotate through multiple representations during the lesson.
Why it works: UDL research shows that providing multiple means of representation improves access and retention for all learners, not just those with identified learning differences. AI handles the content transformation — turning a paragraph into a diagram description, a sequencing activity, or a discussion guide — in minutes.
8. AI-adapted materials for English language learners
The scenario: A 3rd-grade classroom includes five students whose home language is Spanish. The class is reading a short story about community helpers.
The AI differentiation: The teacher uses AI to create:
A bilingual vocabulary list with key terms in English and Spanish, simple definitions, and visual cues
A simplified version of the story with shorter sentences and high-frequency vocabulary, keeping the same plot and characters
Comprehension questions in both English and Spanish so students can demonstrate understanding in the language they are most comfortable with
Why it works: English language learners often understand concepts at a higher level than their English proficiency allows them to demonstrate. By providing bilingual supports and simplified text, the teacher ensures these students access the same content without being held back by language barriers. The SAMR model places this kind of AI use at the Augmentation level — the technology directly improves the teacher's ability to serve multilingual learners without fundamentally redesigning the task.
9. Differentiated scaffolding for students with learning differences
The scenario: A high school English class includes students with dyslexia, ADHD, and executive function challenges alongside neurotypical peers. The class is writing an analytical essay about a novel.
The AI differentiation: The teacher generates tailored scaffolds for different needs:
For students with dyslexia: A graphic organizer with sentence starters, bullet-point structure instead of full paragraphs for the planning stage, and a simplified rubric with examples of what "meets expectations" looks like
For students with ADHD: The essay broken into five short, timed tasks (10 minutes each) with a checklist and clear stopping points
For students with executive function challenges: A step-by-step planning template that walks through thesis → evidence → analysis → conclusion, with a model paragraph for reference
For advanced writers: A constraints-based challenge — write the essay from an opposing viewpoint, or limit the analysis to only subtext and implied meaning
Why it works: Each student writes an analytical essay about the same novel, meeting the same learning standard. But the scaffolding accounts for how different brains process and organize information. A 2024 UK Department for Education survey found that 57% of special education teachers already use AI tools for creating adapted resources, making this one of the fastest-growing applications of AI in education.
10. Adaptive questioning for retrieval practice
The scenario: A middle school science teacher wants to start every class with a five-minute retrieval practice warm-up, but a single set of questions is always too easy for some students and too hard for others.
The AI differentiation: The teacher prompts AI to generate three question sets from the previous week's content on cells and cell organelles:
Recall questions: "What is the function of the mitochondria?" / "Name three organelles found in both plant and animal cells."
Application questions: "A cell is not producing enough energy. Which organelle might be malfunctioning, and why?"
Transfer questions: "If you were designing a new type of cell that needed to survive in extreme heat, which organelles would you modify and how?"
The teacher projects all three sets and lets students choose their level, or assigns levels based on the previous day's exit ticket data.
Why it works: Retrieval practice is one of the most evidence-backed learning strategies, supported by decades of cognitive science research. But it only works when the retrieval demand matches the student's current level of knowledge. Too easy, and there is no retrieval effort. Too hard, and the student guesses rather than retrieves. AI-generated tiered questions ensure every student is in the productive struggle zone. This approach maps directly onto Rosenshine's Principles of Instruction, particularly Principle 1 (daily review) and Principle 6 (checking for understanding).
How to get started with AI-differentiated instruction
If these examples feel overwhelming, start small. Choose one lesson this week with a class you know well. Pick a single activity you would normally give to everyone as the same version. Then follow this framework:
Identify the learning objective. What should every student know or be able to do by the end of the lesson?
Choose your differentiation type. Are you differentiating by readiness, interest, or learning profile?
Write a specific AI prompt. Include the grade level, subject, learning objective, number of versions, and what changes between versions. The more specific your prompt, the better the output.
Review and adjust. AI-generated resources are drafts, not finished products. Spend five minutes checking vocabulary, accuracy, and alignment with your curriculum.
Distribute and observe. Assign versions based on your professional judgment, not fixed labels. Watch how students respond and adjust for the next lesson.
TeacherPlug makes this process even smoother. Instead of starting from scratch with AI prompting, TeacherPlug provides structured tutorials that walk you through differentiation workflows step by step, a curated prompt library organized by subject, grade level, and task type, and material generators that produce classroom-ready resources in minutes. Whether you are brand new to AI or already experimenting, TeacherPlug, an AI learning platform for teachers, helps you move from "I should try this" to "I use this every week."
Common mistakes to avoid when using AI for differentiated instruction
Even with AI handling the production work, there are pitfalls to watch out for:
Differentiating down instead of across. When you ask AI to "simplify," it often just removes content. Instead, prompt it to scaffold access to the same learning objective. The support version should provide a different route to the same destination, not a shorter trip to a lesser one.
Static grouping. Just because a student needed the support version for fractions does not mean they need it for geometry. Move students between tiers based on the specific topic, not a permanent label.
Skipping the review. AI makes mistakes. A "simplified" passage might still use vocabulary that is too complex. An extension problem might go beyond the curriculum. Five minutes of teacher review prevents twenty minutes of confusion during the lesson.
Over-differentiating. Three versions is usually enough. Five tiers creates management overhead without meaningful pedagogical benefit.
The future of differentiated instruction is AI-assisted
Differentiated instruction has always been good teaching. The barrier was never the philosophy — it was the time. AI removes that barrier. With the right tools and a clear understanding of your students, you can create genuinely personalized learning experiences without burning out in the process.
The ten examples in this article are starting points, not endpoints. Every classroom is different, and the best differentiation comes from a teacher who knows their students combined with a tool that can execute on that knowledge quickly. If you are looking to master AI tools for your classroom without the overwhelm, TeacherPlug walks you through it step by step — from your first AI-generated worksheet to a fully differentiated unit plan.
.png)