3Class
Class 3 — Foundations of Computational Thinking
At this stage, students are introduced to the four pillars of CT through simple, visual, and observation-based problems involving numbers, shapes, pictures, and everyday situations. All learning is device-free and integrated into Mathematics and TWAU (The World Around Us) using CBSE-provided CT resource books and worksheets. No AI is introduced at this stage.
⏱ 50 Hours/Year
📚 Maths + TWAU
🎲 Puzzles & Activities
🚫 No AI — CT Only
👩🏫 Subject Teachers
📋 Official CBSE Curriculum Statement — Preparatory Stage
"Puzzles, visual reasoning, and problems that require analytical thinking, logical deduction, and basic problem-solving in everyday contexts."— CBSE CT & AI Curriculum Document, Page 10
👩🏫 Who Teaches — Class 3
➕
Mathematics Teacher
Delivers all CT content using the CBSE CT resource book mapped to the Maths textbook. CT questions are chapter-by-chapter.
🌿
TWAU / EVS Teacher
Delivers CT worksheets integrated into The World Around Us topics — classifying living things, sequencing events, observing patterns in nature.
CT Learning Outcomes — Official CBSE Syllabus (Class 3)
💡
Abstract Thinking
3D viewpoints · shape transformations · hidden patterns · incomplete figures
💡 Abstract Thinking
📋 Official Learning Outcome Class 3
- Solve problems with hidden or unseen ideas using different viewpoints of 3D objects
- Identify changes in shapes after flips, turns, folds, or rotations
- Find hidden or missing parts in incomplete shapes or patterns
What This Skill Builds in Students
Abstract thinking develops the ability to mentally manipulate objects — turning, flipping, or rotating them in the mind without physically touching them. This skill allows students to represent real objects as simplified models, which is the first step in how computers process images, maps, and spatial data.
🎯 Classroom Activity — Fold, Cut, and Predict
Fold a paper in half. Draw half a butterfly on the folded edge. Before unfolding, ask students: "What will the shape look like when opened?" Students sketch their prediction. Then unfold and compare. Extend: fold into quarters, cut a small triangle from the corner — now predict before unfolding.
Key discussion: "Each fold is like a rule the computer applies. The result depends on following the rule exactly."
Key discussion: "Each fold is like a rule the computer applies. The result depends on following the rule exactly."
🌍
Real-Life Connection
When you apply a "mirror" filter on Instagram or a photo app, the app is performing the exact same operation students practised — a mathematical reflection transformation applied to every single pixel in the image, thousands of times per second.
🔍
Pattern Recognition
1 or 2 changes per term · numbers, shapes, letters, or mixed patterns
🔍 Pattern Recognition
📋 Official Learning Outcome Class 3
- Identify simple patterns involving 1 or 2 changes in consecutive terms
- Patterns formed using numbers, shapes or images, letters, or a mix of the above
What This Skill Builds
Pattern recognition is the ability to find an underlying rule in a sequence of items. In computing, every repetitive process follows a pattern (a loop). Students who can identify patterns can understand how computers process repetitive instructions efficiently.
🎯 Activity 1 — Number Pattern Discovery
Series: 5, 10, 15, 20, __ · Ask: "What is the rule?" (Add 5 each time.) "What is the 8th term?" (40.) Now extend: "Can you write this as a sentence?" — "Start at 5. Keep adding 5." This sentence is a program instruction — a
for loop.🎯 Activity 2 — Mixed Letter-Number Pattern
Series: A1, B2, C3, D4, __ · Two rules together: letters follow the alphabet (A→B→C), numbers count upward (1→2→3). Ask: "What is the 7th term?" (G7.) Each rule is independent — but they run simultaneously. This is how AI processes multiple data features at the same time.
🌍
Real-Life Connection
Indian vehicle number plates follow a strict pattern: State Code (UP) + District (32) + Letters (AB) + Number (1234). A traffic AI camera reads this pattern in milliseconds and identifies the state, registration district, and vehicle type — pure pattern recognition from a fixed multi-part rule.
🧩
Decomposition
2–3 clue problems · 3D parts · step-by-step transfers · tables with information
🧩 Decomposition
📋 Official Learning Outcome Class 3
- Break down problems involving 2–3 clues using number clues or names
- 3D objects and their parts (faces, edges, corners)
- Step-by-step exchanges or transfers (money, objects, digits)
- Tables or charts with multiple pieces of information
Decomposition means breaking a complex problem into smaller, manageable parts. In computing, every large program is made of small functions — each function solves one small piece. A student who decomposes naturally can understand modular programming and AI system design.
🎯 Activity — Animal Classification Chart (TWAU)
Create a three-column table: Flies | Swims | Walks. Give 15 animal picture cards. Students place each card in the correct column. Key discovery: a duck fits in two columns (Flies AND Swims). A whale fits only in Swims. An eagle fits in Flies AND Walks.
Discussion: "When you search for 'wildlife photos' on Google, the AI has sorted millions of images by labels — exactly like our table. But each image can have multiple labels, just like the duck. This is called multi-label classification."
Discussion: "When you search for 'wildlife photos' on Google, the AI has sorted millions of images by labels — exactly like our table. But each image can have multiple labels, just like the duck. This is called multi-label classification."
🌍
Real-Life Connection
When you search "cat videos" on YouTube, the platform's AI searches through millions of videos labelled with tags like "cat," "pet," "animal," "cute" — exactly like our animal sorting table. Teaching AI to label data correctly is how AI is trained. No labels = no learning.
⚙️
Algorithmic Thinking
Step-by-step rules · grid movement · before/after sequencing · multi-step instructions
⚙️ Algorithmic Thinking
📋 Official Learning Outcome Class 3
- Follow clear step-by-step rules — number sequences using simple operations
- Movements on grids or direction-based paths
- Events arranged using before/after/in-between clues
- Values that increase or decrease across steps
- Multi-step instructions involving moves, changes, or transfers
Algorithmic thinking is the ability to create and follow a precise sequence of steps to solve a problem. Every computer program is an algorithm. A student who thinks algorithmically can move naturally into coding, AI model design, and logical problem-solving across all subjects.
🎯 Activity 1 — Story Scramble (Language Integration)
Give 5 picture cards: Seed → Water → Sprout → Plant → Flower. Students arrange in order. Then ask: "What happens if we skip Step 2 — watering?" The plant dies. The sequence breaks. Key lesson: An algorithm must have every step in the right order. One missing or wrong step = the program fails. This is called a "bug" in programming.
🎯 Activity 2 — Robot on a Grid (Maths Integration)
Draw a 6×6 grid. Place a coin at A1. Give spoken instructions: "Move 3 steps right, then 2 steps down." Ask: "Where does the robot land?" Increase difficulty: "Move right until you reach column 5, then move down 3." This is a directional algorithm — the same logic used in Google Maps, drone navigation, and self-driving vehicle systems.
🌍
Real-Life Connection
When you ask Google Maps "How do I get from Aligarh to Agra?", it runs a pathfinding algorithm — starting at point A, checking every possible turn on a grid of roads, and finding the fastest path. This is the robot-on-a-grid activity, but on a real map with millions of nodes.
📋 Assessment Methods — Class 3 (Official)
Written tests with CT puzzles and visual problems · Interactive group activities (e.g. treasure hunts) · Teacher Observation Journal tracking individual student reasoning process · Focus: thinking process, not just correct answer
4Class
Class 4 — Building Pattern Depth & Structured Logic
CT complexity increases to moderate level. Students work with multi-change patterns, 3D shape properties in more detail, mirror images, symmetry, and more elaborate algorithmic sequences. Still fully device-free, still integrated into Maths and TWAU. Problems now require combining 2–3 clues simultaneously.
⏱ 50 Hours/Year
📚 Maths + TWAU + Language
🔄 Moderate to Complex
🚫 No AI — CT Only
📋 Official CBSE Stage Description
"Puzzles, visual reasoning, and problems that require analytical thinking, logical deduction, and basic problem-solving in everyday contexts — with increasing complexity at Class 4." — CBSE CT & AI Curriculum, Page 10
CT Learning Outcomes — Official CBSE Syllabus (Class 4)
💡
Abstract Thinking
Moderate 3D · mirror images · symmetry · partially visible patterns
📋 Official Learning Outcome Class 4
- Solve moderate to highly moderate problems with partially visible or incomplete ideas
- Different viewpoints of 3D objects; changes after flips, turns, cuts/folds, rotations
- Mirror images and identical halves based on symmetry
🎯 Classroom Activity — Symmetry and Code Reuse
Students fold a paper, draw half a design on the folded edge, cut and unfold. Both halves are mirror images. Ask: "If a computer draws 1,000 of these shapes, does it calculate both halves every time?" No — it calculates one half and uses a mirror function to create the other. This is the programming principle of code reuse — compute once, apply many times — which saves enormous processing power in computer graphics, game engines, and AI image generation.
🌍
Real-Life Connection
Every face in every video game is drawn using mirror symmetry — the game engine only computes one half of the face and reflects it. Without this technique, every frame of a modern video game would require twice the computing power. Symmetry is efficiency.
🔍
Pattern Recognition
One or more changes · numbers, shapes, letters, or mixed patterns
📋 Official Learning Outcome Class 4
- Identify patterns involving one or more changes in consecutive terms
- Patterns using numbers, shapes/images, letters, or a mix
🎯 Activity — Multiplication Table as an Algorithm
Table of 4: 4, 8, 12, 16, 20... Ask: "What is the rule?" (Start at 0, add 4 each time.) "Using only this rule — without memorising — what is 4 × 9?" Students add 4 nine times. Explain: A computer never memorises a multiplication table. It follows the rule: add 4 repeatedly for 9 steps. This is a
for loop — one of the most fundamental structures in any programming language.💡
Teacher Tip
Ask students: "If the rule changes to 'add 7 each time', what changes in the output?" This helps them understand that changing a rule in a program changes all results instantly — introducing the concept of variables in programming.
🧩
Decomposition
Moderate clue clusters · place values · step-by-step transfers · sorting conditions
📋 Official Learning Outcome Class 4
- Break down problems involving a cluster of moderate clues
- Number clues (place values, sum/difference/product); 3D object parts
- Step-by-step exchanges or transfers (money, objects, digits, quantities)
- Conditions for counting, grouping, or sorting items
🎯 Activity — Food Chain as a Flowchart (Science/EVS)
Draw the food chain: Sun → Grass → Grasshopper → Frog → Snake → Eagle. Convert it into a proper flowchart with boxes and arrows. Then ask: "If we remove the frog from this chain, what happens step by step?"
Trace: Grasshopper population explodes → Grass disappears → Snake starves → Eagle finds less food. Students discover that each step depends on the output of the previous step. In programming, this is called a sequential dependency — the same principle that governs how AI weather models compute temperature, humidity, and pressure in a chain before giving a forecast.
Trace: Grasshopper population explodes → Grass disappears → Snake starves → Eagle finds less food. Students discover that each step depends on the output of the previous step. In programming, this is called a sequential dependency — the same principle that governs how AI weather models compute temperature, humidity, and pressure in a chain before giving a forecast.
🌍
Real-Life Connection
In 2022, a global semiconductor shortage disrupted car manufacturing worldwide. Removing one component in the supply chain (decomposition) caused a cascade of problems — exactly like removing the frog from the food chain. Supply chain AI analyses these dependencies to predict shortages before they happen.
⚙️
Algorithmic Thinking + Boolean Logic
Elaborate conditions · grid paths · AND/OR sorting · procedural writing
📋 Official Learning Outcome Class 4
- Follow well-defined, elaborate conditions — moderate to complex problems
- Multi-step instructions involving moves, changes, transfers, swaps
- People/events arranged using attributes or chronological clues
🎯 Activity 1 — Write a Perfect Algorithm (Language)
Students write step-by-step instructions for making a paper boat — without showing each other. Swap. Follow the partner's instructions exactly, with no improvising. Observe what goes wrong. Discuss: "A computer program has zero tolerance for ambiguity. Every step must be perfectly precise and in the right order." An unclear instruction in a program causes it to produce wrong results — this is called a logic error or bug.
🎯 Activity 2 — Boolean Logic Card Sort
Give students 20 coloured shape cards. Ask them to find cards that are:
• Red AND Circular (both conditions must be true)
• Blue OR Triangular (at least one condition true)
• NOT Yellow (condition must be false)
Explain: These are AND, OR, NOT — the three Boolean operators that are the foundation of every computer chip and every database search. When you search "cheap hotels near Agra" — the AI applies: Price = cheap AND Location = Agra AND Type = hotel. Milliseconds. Millions of results.
• Red AND Circular (both conditions must be true)
• Blue OR Triangular (at least one condition true)
• NOT Yellow (condition must be false)
Explain: These are AND, OR, NOT — the three Boolean operators that are the foundation of every computer chip and every database search. When you search "cheap hotels near Agra" — the AI applies: Price = cheap AND Location = Agra AND Type = hotel. Milliseconds. Millions of results.
🌍
Real-Life Connection
Every Google search uses Boolean logic. The query "Taj Mahal" AND "entry fee" NOT "film" finds pages about visiting the monument, not about the film. Hospital database queries: find patients aged over 60 AND diabetic AND NOT on insulin — this saves doctors hours of manual searching.
5Class
Class 5 — Advanced CT Reasoning & Bridge to AI
The most demanding preparatory class. Students handle multi-layered abstract problems, progressive patterns with multiple simultaneous changes, interconnected clue clusters, and advanced multi-step algorithms with conditional logic. By the end of Class 5, every student has all four CT skills at a complex level — fully ready to begin AI in Class 6.
⏱ 50 Hours/Year
🔀 Multi-layered Problems
🔗 Bridge to AI — Class 6
🚫 No AI — CT Only
Official CBSE Curriculum (Page 13): "CT in Classes 3–5 would encourage learners to approach problems in a systematic and logical manner by applying processes such as decomposition, pattern recognition, abstraction, and algorithmic thinking." By Class 5, all four CT skills are developed at a complex level — forming the cognitive foundation for AI literacy in Classes 6–8.
CT Learning Outcomes — Official CBSE Syllabus (Class 5)
💡
Abstract Thinking
Complex multi-layered problems · clockwise/counter-clockwise · water images
📋 Official Learning Outcome Class 5
- Solve complex problems with multi-layered hidden cues
- Changes in shapes — clockwise or counter-clockwise direction changes
- Mirror images AND water images — both applied together
- Hidden or missing parts in complex incomplete patterns
🎯 Activity — Compound Transformation Puzzle
Show a word (e.g. "HELP"). Ask: "What does it look like as a water reflection AND a mirror reflection applied together?" Students must apply two transformations mentally in sequence. This is called a compound transformation. In image processing, photo filters apply dozens of compound transformations per second. In AI image generation, a single prompt ("show me a sunset reflected in water") triggers hundreds of these operations instantly.
🔍
Pattern Recognition
Progressive patterns · multiple simultaneous changes · mixed types
📋 Official Learning Outcome Class 5
- Identify progressive patterns involving multiple simultaneous changes
- Using numbers, shapes/images, letters, or a mix — with multi-variable rules
🎯 Activity — Three-Variable Pattern (Direct AI Connection)
Series: A1+, B4-, C9+, D16-, E25+... Three simultaneous rules: (1) Letters follow alphabet, (2) Numbers are perfect squares (1,4,9,16,25), (3) Signs alternate (+, -, +, -, +). Ask: "What is the 8th term?" (H64-)
Connect directly: An AI weather model combines multiple data patterns simultaneously — temperature is rising (+), humidity is falling (-), pressure is steady. Each variable follows its own rule. The AI finds all three patterns at once and predicts tomorrow's weather. Students just did exactly what the AI does — with three variables.
Connect directly: An AI weather model combines multiple data patterns simultaneously — temperature is rising (+), humidity is falling (-), pressure is steady. Each variable follows its own rule. The AI finds all three patterns at once and predicts tomorrow's weather. Students just did exactly what the AI does — with three variables.
🧠
Did You Know?
The Fibonacci series (1, 1, 2, 3, 5, 8, 13, 21...) appears in sunflower seeds, pinecones, and galaxy spirals. AI researchers use Fibonacci-like growth patterns to optimise search algorithms. Nature and computing share the same mathematical patterns.
⚙️
Algorithmic Thinking — If-Then-Else Logic
Multi-layered conditional rules · cyclic algorithms · if-then reasoning
📋 Official Learning Outcome Class 5
- Follow multi-layered rules to solve advanced problems
- Multi-step instructions involving moves, changes, transfers, swaps
- Complex ordering using multiple attributes and chronological clues
🎯 Activity 1 — If-Then-Else Grammar (Language Integration)
Conditional sentences in grammar are identical in structure to if-then statements in programming. Complete these:
• If it is raining, then ________, else ________.
• If today is a holiday, then ________, else ________.
• If score ≥ 90, then grade = A, else if score ≥ 75, then grade = B, else grade = C.
Explain: Every second, your smartphone evaluates thousands of if-then conditions. "If battery < 15%, then dim screen AND send low battery alert." "If you are moving at 60km/h, then assume you are in a vehicle AND update GPS." If-then-else = the most fundamental decision-making structure in all programming.
• If it is raining, then ________, else ________.
• If today is a holiday, then ________, else ________.
• If score ≥ 90, then grade = A, else if score ≥ 75, then grade = B, else grade = C.
Explain: Every second, your smartphone evaluates thousands of if-then conditions. "If battery < 15%, then dim screen AND send low battery alert." "If you are moving at 60km/h, then assume you are in a vehicle AND update GPS." If-then-else = the most fundamental decision-making structure in all programming.
🎯 Activity 2 — Life Cycle as a Cyclic Algorithm (Science)
Draw the butterfly life cycle as a flowchart: 🥚 Egg → 🐛 Caterpillar → 🫘 Cocoon → 🦋 Butterfly → back to 🥚 Egg.
Point out that the last arrow returns to the start. Ask: "When does this cycle stop?" (When the butterfly cannot reproduce — the condition becomes false.) In programming:
This is a while loop — repeat a sequence as long as a condition is true. One of the most important structures in any programming language.
Point out that the last arrow returns to the start. Ask: "When does this cycle stop?" (When the butterfly cannot reproduce — the condition becomes false.) In programming:
while species_alive: complete_lifecycle()This is a while loop — repeat a sequence as long as a condition is true. One of the most important structures in any programming language.
🌍
Real-Life Connection
Your phone's autocorrect uses a conditional chain: "If the typed word is NOT in the dictionary, AND it sounds like a known word (edit distance ≤ 2), AND the next letter suggests the known word, THEN suggest the correction." Three conditions evaluated simultaneously — millions of times per day. This is Class 5 logic running on your phone.
📋 Assessment Methods — Classes 3–5 (Official)
Written tests with CT puzzles · Interactive group activities · Teacher Observation Journal · Competency-based: assess ability to apply and reason, not to recall or memorise
6Class
Class 6 — Advanced CT Continues + First Steps into AI
CT skills advance significantly — problems are now complex, multi-layered, and require combining CT skills together. Simultaneously, AI is formally introduced for the first time: what it is, how it differs from automation, basic data concepts, digital ethics. Subject teachers handle CT; Computer teacher handles AI. Both collaborate on the 40-hour project component.
⏱ 100 Hours/Year Total
🧠 40 hrs Advanced CT
🤖 20 hrs AI Literacy
📐 40 hrs Projects
🖥️ Computer Lab Required
📋 Official CBSE Curriculum Statement — Middle Stage (Classes 6–8)
"Advanced CT, data organisation, representation and analysis, AI awareness, and ethical aspects of AI."— CBSE CT & AI Curriculum Document, Page 10
40
Advanced CT
Complex CT aligned to Maths textbook · Subject teachers
20
AI Literacy
AI intro, data, ethics · Computer teacher
40
Projects
Cross-subject CT+AI · Assessed by Computer teacher
👩🏫 Who Teaches What — Classes 6–8 (Official CBSE)
📚
Subject Teachers (Maths, Science, Social Science, Language)
Deliver Advanced CT content using CT resource books mapped to the Class textbook. Questions are more challenging than Classes 3–5 and require combining multiple CT skills.
🖥️
Computer Teacher
Delivers all AI Literacy content (20 hrs) using the AI Foundation Handbook. Also assesses all interdisciplinary project work.
🤝
Collaborative — Subject Teacher + Computer Teacher
The 40-hour project component requires both teachers working together. Subject teachers contribute CT aspects; Computer teacher provides AI tools and frameworks. Projects integrate Maths, Science, Social Studies, and Language.
🧠
Advanced CT Module — Class 6 (40 Hours)
40 hrs
Advanced Abstract Thinking
Compound transformations, symmetry across multiple axes, visual reasoning with scale and proportion, advanced 3D cross-sections
Advanced Pattern Recognition
Patterns with alternation, skipping, grouping, cyclic behaviour — mixed patterns with dependency rules between variables
Advanced Decomposition
Interdependent clues, factors, multiples, 2D/3D shape properties, grids requiring cross-referencing of multiple data points
Advanced Algorithmic Thinking
Multi-layered rules, combined operations, logical flow — identifying necessary vs redundant information in a procedure
🤖
AI Literacy Module — Class 6 (20 Hours) — Computer Teacher
20 hrs
| # | Topic | Hrs | Key Learning Points |
|---|---|---|---|
| 1 | Introduction to AI & Everyday Examples | 5 | What AI is; AI vs Automation; human vs machine intelligence; supervised, unsupervised, and reinforcement learning |
| 2 | Basic Data Concepts | 5 | Data types: numbers, text, images, sound; organising and representing data in tables and charts |
| 3 | Simple Pattern Recognition & Decision Making | 5 | Identifying patterns in data; making simple decisions based on observations and data |
| 4 | Ethics and Digital Responsibility | 5 | Online safety, passwords, privacy, digital footprints, ethical and responsible use of technology |
Subject-wise CT + AI Integration — Class 6
🔢
Mathematics — Data Organisation & AI Data Connection
Advanced CT: complex data tables · AI: how structured data powers AI predictions
💡 Advanced Abstract🔍 Advanced Pattern📊 Data Organisation
📋 Official CT Learning Outcome Class 6
- Identify, extend, and justify complex patterns with multiple simultaneous changes
- Patterns involving alternation, skipping, grouping, or cyclic behaviour
- Logical flow of steps — identifying necessary vs redundant information
Advanced CT Activity
🎯 Activity — Organise the Attendance Dataset
Give students 30 random daily attendance numbers (e.g. 42, 38, 45, 29, 41...). Ask them to: (1) Sort them in ascending order, (2) Group them by week in a table, (3) Calculate the weekly average, (4) Identify the week with highest absence, (5) State why this pattern might exist.
Each step is a distinct CT skill: sorting = algorithmic thinking, grouping = decomposition, finding the pattern = pattern recognition, ignoring individual names and focusing on totals = abstraction. Students have just performed data analysis — the core skill of a data scientist.
Each step is a distinct CT skill: sorting = algorithmic thinking, grouping = decomposition, finding the pattern = pattern recognition, ignoring individual names and focusing on totals = abstraction. Students have just performed data analysis — the core skill of a data scientist.
AI Connection
🤖 AI Activity — What Could an AI Do With This Data?
After completing the attendance table, ask: "If an AI had access to 3 years of this data, what could it predict or detect automatically?"
Student responses to guide: Which month has most absences (perhaps cold season or exam stress), which student is at risk of failing due to low attendance, whether Mondays consistently have higher absences, whether a particular teacher's class has higher engagement.
Explain: This is exactly how school management AI software works — it reads structured data tables, identifies statistical patterns, and sends automatic alerts to teachers. No magic — just organised data + pattern detection.
Student responses to guide: Which month has most absences (perhaps cold season or exam stress), which student is at risk of failing due to low attendance, whether Mondays consistently have higher absences, whether a particular teacher's class has higher engagement.
Explain: This is exactly how school management AI software works — it reads structured data tables, identifies statistical patterns, and sends automatic alerts to teachers. No magic — just organised data + pattern detection.
🌍
Real-Life Connection
Swiggy and Zomato maintain data tables exactly like your attendance sheet — but with millions of rows: user ID, dish ordered, time, location, day, weather. Their AI reads these tables, finds patterns (e.g., you order biryani on Fridays), and shows it first on your home screen the following Friday. Your attendance table is doing what a ₹1,000-crore company does at scale.
🔬
Science — Classification Trees & Supervised Learning
Advanced CT: decision trees · AI: how machines learn from labelled examples
🧩 Advanced Decomposition🤖 Supervised Learning
🎯 CT + AI Activity — Build a Decision Tree
Students build a Yes/No classification tree to identify 15 living organisms:
Is it alive? → Yes → Does it have a backbone? → Yes → Does it have feathers? → Yes → It is a Bird.
Build a complete tree covering all 15 organisms. Each Yes/No question is a node. The final identification is a leaf.
Key insight: This hand-drawn tree is literally the diagram of a Decision Tree Algorithm — one of the most important and widely-used algorithms in machine learning. Medical diagnosis AI, loan approval systems, and spam filters all use decision trees built from thousands of labelled examples.
Is it alive? → Yes → Does it have a backbone? → Yes → Does it have feathers? → Yes → It is a Bird.
Build a complete tree covering all 15 organisms. Each Yes/No question is a node. The final identification is a leaf.
Key insight: This hand-drawn tree is literally the diagram of a Decision Tree Algorithm — one of the most important and widely-used algorithms in machine learning. Medical diagnosis AI, loan approval systems, and spam filters all use decision trees built from thousands of labelled examples.
🌍
Real-Life Connection
Open Google Lens and photograph a leaf — it gives you the plant name in 2 seconds. The AI ran your photo through a decision tree with millions of nodes: shape of leaf edge → vein pattern → colour → texture → match. The tree was built from 50 million labelled plant photos contributed by botanists worldwide. You built a 15-node tree. Google's has millions.
🏛️
Social Science — Digital Footprints & Privacy Ethics
AI Ethics: what data you leave online and how AI uses it
🛡️ Digital Ethics
🤖 Activity — Map Your Digital Footprint
Students list every app/website used in the past 24 hours. For each, discuss: what data does it collect?
• WhatsApp: messages, contacts, location, device model, usage time
• YouTube: videos watched, duration, search history, device, location
• Google Maps: every location you visited, time spent, route taken
• Phone Camera: photo GPS coordinates, time, device metadata
Draw it as a large footprint diagram with data types written inside each toe.
Explain: This entire footprint is used by AI systems to build a profile of who you are, what you like, where you go, and even predict your future behaviour. This is why a privacy setting exists — controlling what data you allow AI to have about you.
• WhatsApp: messages, contacts, location, device model, usage time
• YouTube: videos watched, duration, search history, device, location
• Google Maps: every location you visited, time spent, route taken
• Phone Camera: photo GPS coordinates, time, device metadata
Draw it as a large footprint diagram with data types written inside each toe.
Explain: This entire footprint is used by AI systems to build a profile of who you are, what you like, where you go, and even predict your future behaviour. This is why a privacy setting exists — controlling what data you allow AI to have about you.
🧠
Did You Know?
In 2020, a major study found that a person's location data from just 4 random points in a week was enough for an AI to identify them with 95% accuracy — even if all names and phone numbers were removed. Data privacy is not optional — it is a fundamental right.
7Class
Class 7 — Data Representation, Analysis & AI Domains
This class directly addresses the official CBSE Middle Stage mandate: "data organisation, representation and analysis, AI awareness, and ethical aspects of AI." CT reaches algebraic and spatial complexity. AI expands to Computer Vision, NLP, and Data Science. Students begin working with real data visualisations and critically evaluate AI bias.
⏱ 100 Hours/Year
🧠 40 hrs Advanced CT
🤖 20 hrs AI Domains
📐 40 hrs "My School in Data"
📋 Official CBSE Statement — Middle Stage Focus
"Advanced CT, data organisation, representation and analysis, AI awareness, and ethical aspects of AI."— CBSE CT & AI Curriculum, Page 10. Class 7 is where this becomes most visible — data visualisation, AI domain awareness, and AI bias ethics are all Class 7 topics.
40
Advanced CT
Algebraic patterns · spatial reasoning · pathfinding algorithms
20
AI Domains
CV · NLP · Data Science · Bias · Industries
40
Data Project
"My School in Data" — collect, visualise, analyse, present
🧠
Advanced CT Module — Class 7 (40 Hours)
40 hrs
Abstract Thinking
Compound transformations — multiple flips + turns + folds combined; symmetry, congruence, proportional reasoning; relative positions and viewpoints
Pattern Recognition
Algebraic patterns using variables and expressions; geometric growth rules; nested and dependent patterns; functional relationships
Decomposition
Spatial reasoning + geometry; tables and charts with multiple dependencies; translating visual/verbal information into structured data for analysis
Algorithmic Thinking
Pathfinding with decision points and constraints; if-then reasoning; elimination strategies; optimality analysis of procedure steps
🧠
AI Literacy Module — Class 7 (20 Hours) — Computer Teacher
20 hrs
| # | Topic | Hrs | Key Learning Points |
|---|---|---|---|
| 1 | AI Domains | 5 | Classification, Regression, Clustering; Computer Vision, NLP, Data Science; hands-on practice applying techniques to a small dataset using AI tools |
| 2 | AI in Industries | 5 | Healthcare, education, transport, communication — how AI improves accuracy, efficiency, and productivity with real examples |
| 3 | Data Visualisation & Analysis | 5 | Collecting structured data; creating bar charts, line graphs, pie charts; interpreting patterns and trends in real datasets |
| 4 | Ethics & AI Bias Awareness | 5 | What bias in AI means; case examples of biased AI causing real harm; responsible use; informed consent before data is collected or shared |
Subject-wise CT + AI Integration — Class 7
📊
Mathematics — Data Representation & Trend Analysis
Advanced CT: anomaly detection · AI: time-series forecasting with real data
📋 Official AI Learning Outcome Class 7 — Data Visualisation
- Collect and organise simple structured data; interpret patterns and trends
- Create bar charts, line graphs, and pie charts from real datasets
- Apply basic predictive approaches to a small dataset
🎯 Activity — Anomaly Detection in Monthly Data
Monthly sales data: 45, 48, 52, 49, 51, 55, 47, 98, 50, 53, 49, 52. Plot a line graph. Ask: "Which value is clearly unusual? Why might that month be different from all others?" (Month 8 = 98 — likely a Diwali sale or special offer.)
Finding values that don't fit the pattern is called outlier or anomaly detection. AI fraud detection systems work exactly this way: if you normally spend ₹500/day and suddenly a transaction of ₹50,000 appears from another city, the AI flags it as a potential fraud — because it's an anomaly in your personal data pattern.
Finding values that don't fit the pattern is called outlier or anomaly detection. AI fraud detection systems work exactly this way: if you normally spend ₹500/day and suddenly a transaction of ₹50,000 appears from another city, the AI flags it as a potential fraud — because it's an anomaly in your personal data pattern.
🤖 AI Activity — Rainfall Prediction from Historical Data
Plot monthly rainfall data for a city (Jan–Dec) as a line graph. Ask:
• "Which months have the highest rainfall?" (Pattern recognition)
• "Based on this 1-year pattern, predict next July's rainfall." (Extending pattern = basic prediction)
• "How could an AI use 30 years of this same data to predict flood risk?"
Explain: The AI identifies the same pattern you identified, but across 360 data points instead of 12 — giving far more confidence in its prediction. This is called time-series forecasting — the technique behind weather apps, agricultural yield prediction, and city flood management AI systems.
• "Which months have the highest rainfall?" (Pattern recognition)
• "Based on this 1-year pattern, predict next July's rainfall." (Extending pattern = basic prediction)
• "How could an AI use 30 years of this same data to predict flood risk?"
Explain: The AI identifies the same pattern you identified, but across 360 data points instead of 12 — giving far more confidence in its prediction. This is called time-series forecasting — the technique behind weather apps, agricultural yield prediction, and city flood management AI systems.
🌍
Real-Life Connection
IMD (India Meteorological Department) uses AI that analyses 70+ years of daily rainfall, temperature, and pressure data from 500+ stations across India. When you see "78% chance of rain" on your weather app, that percentage came from an AI that found how often it rained under exactly these conditions in historical data. Your line graph activity is the foundation of this system.
🏥
Science — Computer Vision in Healthcare
AI Domain: how machines see and interpret images — diagnosis AI examples
🤖 Computer Vision
📋 Official AI Learning Outcome Class 7 — AI Domains
- Computer Vision: basics of how machines understand and respond to visual information
- AI in healthcare: how AI improves accuracy and efficiency of diagnosis
🤖 Activity — AI vs Human Diagnosis Discussion
Show students two chest X-ray images side by side — one healthy, one with early-stage pneumonia. Ask them to identify the difference. (Most students will struggle — the differences are subtle.)
Explain: An AI model trained on over 100,000 labelled X-ray images can detect early-stage pneumonia in under one second with very high accuracy.
Discussion questions:
• Does this replace doctors or assist them? (It assists — doctors make the final decision)
• What if the AI was trained only on X-rays from patients in the USA? (Might not perform as well on Indian patients with different average body structures)
• What happens when the AI is wrong and a patient goes untreated? (This is why human oversight is always required)
Explain: An AI model trained on over 100,000 labelled X-ray images can detect early-stage pneumonia in under one second with very high accuracy.
Discussion questions:
• Does this replace doctors or assist them? (It assists — doctors make the final decision)
• What if the AI was trained only on X-rays from patients in the USA? (Might not perform as well on Indian patients with different average body structures)
• What happens when the AI is wrong and a patient goes untreated? (This is why human oversight is always required)
🌍
Real-Life Connection
During the COVID-19 pandemic (2020–21), several Indian hospitals used AI tools to analyse CT scans and identify infected lung tissue — helping overloaded doctors manage hundreds of patients daily when human capacity was severely stretched. AI as a tool did not replace the doctor — it extended what one doctor could review in a day from 30 scans to 300 scans.
⚖️
Social Science — Ethical Aspects of AI: Understanding Bias
Official CBSE focus: "ethical aspects of AI" — bias, fairness, digital citizenship
⚖️ AI Ethics🛡️ AI Bias
📋 Official AI Learning Outcome Class 7 — Ethics & Bias
- Explain what bias in AI means and identify situations where AI gives unfair results
- Use safe practices for data privacy — informed consent before data is collected, used, shared, or deleted
- Demonstrate responsible use of technology as part of good digital citizenship
🤖 Case Discussion — The Biased Traffic AI
Scenario: A smart city AI traffic management system is installed in a medium-sized town in Rajasthan. It was trained entirely on road and traffic data from Delhi's highway network. After installation, the AI constantly recommends routes that do not exist in this smaller town, causing traffic jams instead of reducing them.
Why did the AI fail? The training data (Delhi highways) did not represent the new environment (smaller Rajasthan town with different road types, widths, and traffic patterns).
What is AI Bias? When an AI's training data does not represent all the users, environments, or situations where the AI will be used — causing it to perform well for some groups and poorly for others.
Real documented example: Facial recognition AI systems have been shown to misidentify darker-skinned faces at significantly higher rates than lighter-skinned faces — because early training datasets had far more images of lighter-skinned people. This has led to incorrect arrests in some cases.
Why did the AI fail? The training data (Delhi highways) did not represent the new environment (smaller Rajasthan town with different road types, widths, and traffic patterns).
What is AI Bias? When an AI's training data does not represent all the users, environments, or situations where the AI will be used — causing it to perform well for some groups and poorly for others.
Real documented example: Facial recognition AI systems have been shown to misidentify darker-skinned faces at significantly higher rates than lighter-skinned faces — because early training datasets had far more images of lighter-skinned people. This has led to incorrect arrests in some cases.
💡
Teacher Discussion Prompt
Ask students: "If an AI loan approval system in India was trained mostly on data from urban applicants, would it be fair to rural applicants? What might it get wrong?" This connects AI ethics directly to students' lived reality in India.
💬
Language — Natural Language Processing: AI & Language
How AI understands human language — and where it fundamentally fails
🤖 NLP
🤖 Activity — Testing Translation AI Limits
Translate three sentences from Hindi to English using Google Translate and evaluate accuracy:
1. "वह स्कूल जाती है।" — Simple sentence (AI usually handles correctly)
2. "कल मैं बाज़ार गया था।" — Complex past tense (AI usually handles)
3. "जो होता है, अच्छे के लिए होता है।" — Cultural idiom (AI usually fails or translates literally)
Discuss: Why does AI struggle with idioms and cultural sayings?
Because AI learns language from billions of text samples — finding word-to-word statistical patterns. But idioms are not literal — "it's raining cats and dogs" has nothing to do with cats or dogs. AI has no cultural understanding, only statistical word association. This is a fundamental limitation of current NLP technology called lack of pragmatic understanding.
1. "वह स्कूल जाती है।" — Simple sentence (AI usually handles correctly)
2. "कल मैं बाज़ार गया था।" — Complex past tense (AI usually handles)
3. "जो होता है, अच्छे के लिए होता है।" — Cultural idiom (AI usually fails or translates literally)
Discuss: Why does AI struggle with idioms and cultural sayings?
Because AI learns language from billions of text samples — finding word-to-word statistical patterns. But idioms are not literal — "it's raining cats and dogs" has nothing to do with cats or dogs. AI has no cultural understanding, only statistical word association. This is a fundamental limitation of current NLP technology called lack of pragmatic understanding.
🌍
Real-Life Connection
India has 22 official languages and hundreds of dialects. Building AI translation tools for regional Indian languages is one of the most important and challenging problems for Indian AI researchers today. Projects like Bhashini (Government of India's AI translation initiative) are attempting to build AI that understands not just words, but cultural context in Indian languages.
8Class
Class 8 — CT Mastery + Full AI Project Lifecycle & Responsible AI
The capstone of the entire 3–8 CT and AI curriculum. CT reaches its highest complexity — binary number systems, optimisation problems, complex conditional decision-making. AI moves from awareness to practice: students build real AI tools using no-code platforms, experience the complete AI project lifecycle, and evaluate AI for fairness, ethics, and responsible use.
⏱ 100 Hours/Year
🧠 40 hrs CT Mastery
🚀 20 hrs AI Lifecycle
🏆 40 hrs Capstone
⚖️ Responsible AI Focus
📋 Official CBSE Statement — Class 8 AI Lifecycle
"Describe the stages of the AI project cycle: Define Problem, Collect Data, Test AI Tools, Reflect and Improve. Apply no-code tools to tackle real-world problems and reflect on their utility and effectiveness." — CBSE CT & AI Curriculum, Page 29
40
CT Mastery
Binary numbers · optimisation · complex conditions
20
AI Lifecycle
Build real AI tools · no-code platforms
40
Capstone
End-to-end AI project with ethics review
🧠
CT Mastery Module — Class 8 (40 Hours)
40 hrs
Abstract Thinking
Number systems: decimal, binary, ternary, Roman, Chinese numerals; logical interpretation of codes and symbols; identifying essential vs irrelevant data
Pattern Recognition
Powers, exponents, and numerical structures; geometric configurations; conditional patterns based on rules, constraints, and dependencies
Decomposition
High-order logical problems; networks and diagrams with multiple dependencies; structuring problems with multiple variables and cases
Algorithmic Thinking
Conditional instructions (if-then, either-or, must/must not); sequential decision-making under limitations; optimisation: maximum or minimum outcomes
🚀
AI Project Lifecycle Module — Class 8 (20 Hours) — Computer Teacher
20 hrs
| # | Topic | Hrs | Key Learning Points |
|---|---|---|---|
| 1 | AI Project Lifecycle (Conceptual) | 5 | Stages: Define Problem → Collect Data → Test AI Tools → Reflect and Improve; how AI learns from data patterns |
| 2 | AI Applications — Deep Dive | 5 | AI in environment, healthcare, automation, education; hands-on experience with no-code AI tools — image classifiers, chatbots, data prediction apps |
| 3 | Data and Fairness | 5 | How AI uses data; identifying bias in datasets; applying basic strategies to ensure fairness and inclusivity in AI systems |
| 4 | Ethics and Responsible AI | 5 | Privacy issues, misinformation, social impact of AI; responsible use of digital tools; reflection on real-world AI challenges |
Subject-wise CT + AI Integration — Class 8
📐
Mathematics — Binary Numbers & Optimisation + Probability as AI Prediction
CT: number systems, algorithm efficiency · AI: probability → prediction models
📋 Official CT Learning Outcome Class 8 — Abstract Thinking
- Generalisation across different number systems: decimal, binary, ternary, Roman, Chinese numerals
- Optimisation problems involving maximum or minimum outcomes
- Sequential decision-making under given limitations
🎯 CT Activity — Binary: The Language of Computers
Introduce: All computers store everything — photos, music, documents, videos — using only two digits: 0 (OFF) and 1 (ON).
Practice: Count in binary: 0, 1, 10, 11, 100, 101, 110, 111, 1000...
Convert: decimal 13 → binary 1101 · decimal 25 → binary 11001
Reverse: binary 1010 → decimal 10 · binary 10011 → decimal 19
Ask: "Why does a computer use binary instead of our decimal system?" Because electronic circuits have exactly two stable states — electricity flowing (1) or not flowing (0). Binary is the natural language of electricity and therefore the natural language of all computing.
Practice: Count in binary: 0, 1, 10, 11, 100, 101, 110, 111, 1000...
Convert: decimal 13 → binary 1101 · decimal 25 → binary 11001
Reverse: binary 1010 → decimal 10 · binary 10011 → decimal 19
Ask: "Why does a computer use binary instead of our decimal system?" Because electronic circuits have exactly two stable states — electricity flowing (1) or not flowing (0). Binary is the natural language of electricity and therefore the natural language of all computing.
🤖 AI Activity — Probability as the Language of AI Prediction
Flip a coin 60 times. Record results in order. Calculate: after two consecutive Heads (HH), how often did the next flip come up Heads in your data? Calculate this probability.
Now use your calculated probability to predict the next flip after HH in a new series.
Discuss: "This is exactly how an AI prediction model works." It analyses historical patterns and calculates: "Given past conditions, what is the probability of each outcome next time?" The weather app says "73% chance of rain" because: in historical data, on days with exactly these temperature + humidity + pressure conditions, it rained on 73 out of every 100 next days. Probability is the mathematical language that allows AI to quantify uncertainty and make predictions.
Now use your calculated probability to predict the next flip after HH in a new series.
Discuss: "This is exactly how an AI prediction model works." It analyses historical patterns and calculates: "Given past conditions, what is the probability of each outcome next time?" The weather app says "73% chance of rain" because: in historical data, on days with exactly these temperature + humidity + pressure conditions, it rained on 73 out of every 100 next days. Probability is the mathematical language that allows AI to quantify uncertainty and make predictions.
🌍
Real-Life Connection
Ola and Uber use probability-based AI to predict where surge demand will occur before it happens — positioning drivers in advance. The AI calculates: "There is a 91% probability of high demand near this metro station at 6pm on a Friday" — based on years of historical ride data. Probability + historical data = AI prediction that saves time and money for millions of people every day.
♻️
Science — Hands-On AI: Build a Waste Classifier
Official CBSE: "hands-on experience with no-code AI tools (image classifiers)" — Teachable Machine
📋 Official AI Learning Outcome Class 8
- Describe stages of the AI project cycle: Define Problem, Collect Data, Test AI Tools, Reflect and Improve
- Apply no-code tools to tackle real-world problems and reflect on their utility and effectiveness
- Explain how AI uses data; find and research sources of bias in datasets
🤖 Official Hands-On Activity — Waste Sorting AI (Google Teachable Machine)
- 1Define Problem: "Build an AI model that can identify whether a waste item is plastic or paper — to help automate waste sorting in schools."
- 2Collect Data: Use any school device camera. Take 30 varied photos of plastic items (bottles, covers, bags — different colours, angles, lighting) and 30 varied photos of paper items (newspapers, cardboard, notebooks). Variety in training data = better AI.
- 3Label & Train: Upload to teachablemachine.withgoogle.com (completely free, browser-based, no account needed). Label each group. Click "Train Model." Watch accuracy improve in real time.
- 4Test AI Tools: Show new items to the camera. Record: how many correct? How many wrong? What types of items confuse the AI most?
- 5Reflect and Improve: Why did it fail on certain items? Lighting too dark? Too similar to other class? Add 20 more photos of the difficult items and retrain. Compare Version 1 accuracy vs Version 2 accuracy.
- 6Ethics Check: Could this AI make an error that causes harm? (E.g., misclassify a biodegradable item as plastic → sent to wrong waste stream.) Who is responsible for fixing this?
Tool: Google Teachable Machine · teachablemachine.withgoogle.com · Completely free · Works in any browser · No downloads, no accounts, no coding · Supports image, audio, and pose classification
⚖️
Social Science — Responsible AI: Real Ethics Cases
How AI bias causes unfair outcomes — accountability, privacy, and serving human interests
📋 Official AI Learning Outcome Class 8 — Ethics
- Recognise how bias in AI leads to unfair conclusions
- Realise the importance of accountability, privacy, and serving human interests
- Describe AI ethics as the values and guidelines ensuring AI is created and used responsibly
- Analyse contributions of AI — understanding both benefits and risks
🤖 Case Study — The Biased Hiring Algorithm (Amazon, 2018 — Real Event)
A major technology company built an AI system to automatically screen thousands of job applications. The AI was trained on 10 years of the company's own past hiring decisions — decades of data on who was hired and who was rejected.
After testing, the company's own engineers discovered the AI was consistently rating female applicants lower for technical positions. It was penalising CVs that included the word "women's" (e.g., "women's robotics team captain", "women's coding club") and downgrading graduates from all-women's colleges.
Why did this happen? In the past decade of training data, men had been hired in technical roles far more frequently than women — reflecting historical industry patterns. The AI learned, statistically, that being male was a positive signal for technical jobs. It wasn't programmed with bias — it discovered bias in historical data and reproduced it.
Who is responsible? The company — AI cannot be held legally or morally responsible. Humans who design, train, and deploy AI systems are accountable for their outcomes.
How was it fixed? They added filters to block gender-correlated features from the training data. But the bias had spread too widely through the model, and the company ultimately shut down the system entirely in 2018.
Core lesson: AI is only as fair as the data it is trained on. Biased data → Biased AI → Unfair outcomes for real people.
After testing, the company's own engineers discovered the AI was consistently rating female applicants lower for technical positions. It was penalising CVs that included the word "women's" (e.g., "women's robotics team captain", "women's coding club") and downgrading graduates from all-women's colleges.
Why did this happen? In the past decade of training data, men had been hired in technical roles far more frequently than women — reflecting historical industry patterns. The AI learned, statistically, that being male was a positive signal for technical jobs. It wasn't programmed with bias — it discovered bias in historical data and reproduced it.
Who is responsible? The company — AI cannot be held legally or morally responsible. Humans who design, train, and deploy AI systems are accountable for their outcomes.
How was it fixed? They added filters to block gender-correlated features from the training data. But the bias had spread too widely through the model, and the company ultimately shut down the system entirely in 2018.
Core lesson: AI is only as fair as the data it is trained on. Biased data → Biased AI → Unfair outcomes for real people.
🧠
India Connection
In India, AI-based loan approval and credit scoring systems are being adopted rapidly. Questions are being raised: do these systems inadvertently penalise applicants from certain states, communities, or income brackets — because historical loan data reflects existing inequalities? Understanding AI bias is not just academic — it is directly relevant to every Indian citizen's financial future.
🏆
Capstone Project — 40 Hours — End-to-End AI Project
Define Problem → Collect Data → Build → Test → Reflect and Improve → Ethics Review → Present
Official CBSE Curriculum (Page 22): "Interdisciplinary projects encourage students to use their understanding of CT and AI to analyse problems, draw connections across disciplines, and design creative, data-driven solutions... Each activity is curated to make learning engaging and hands-on. Through these experiences, students strengthen CT and AI skills, develop critical thinking, and appreciate the interconnected nature of knowledge."
🎯 Sample Capstone Project — School FAQ Chatbot
- 1Define Problem (CT — Decomposition): "New students joining Class 6 do not know school timings, rules, clubs, exam schedules, and canteen menus. Build a chatbot that can answer their top 50 questions automatically, available 24/7."
- 2Collect Data (CT — Abstraction): Survey Class 6 students — what were their top questions when they joined? Write 50 clear question-and-answer pairs. This is your training dataset. Quality and coverage of this data determines the quality of your AI.
- 3Test AI Tools: Use a free no-code chatbot platform. Enter your Q&A pairs as the knowledge base. No programming needed.
- 4Reflect — Version 1 Testing: Ask 5 classmates who did not help build it to use the chatbot freely for 10 minutes. Record every question it answered incorrectly or did not understand.
- 5Improve — Version 2: Add the missing Q&A pairs. Rewrite unclear answers. Test again with a different group. Document Version 1 accuracy vs Version 2 accuracy — what percentage of questions did it answer correctly in each version?
- 6Ethics Review: Does the chatbot collect any user data? If yes, is it stored securely? Does it give any factually incorrect information? How will errors be reported and fixed in future? Who is responsible if the chatbot gives wrong information?
- 7Present (10 minutes): Cover: Problem Statement, Dataset Description, Tool Used, Version 1 Results, What Changed in Version 2, Limitations, and Ethical Considerations. Assessed by Computer teacher using official rubric.
📋 Official Assessment — Classes 6–8 (CBSE Curriculum, Page 18)
Written tests · Interactive group activities · Practical examinations · Teacher Observation Journal · Thematic projects assessed by Computer teacher · Reflective journals · Group discussions and debates on ethical dilemmas in AI use · Assessment focus: ability to apply knowledge and design solutions, NOT rote memorisation of facts