Class 12 Artificial Intelligence (843) – CBSE Syllabus 2026–27
Complete curriculum guide and examination blueprint for CBSE Class 12 Artificial Intelligence (Subject Code 843, Job Role: AI Assistant) for session 2026–27. Master Data Science Methodology, Computer Vision, Orange Data Mining, Big Data Analytics, Neural Networks, Generative AI, Data Storytelling, and 50-mark Practical & Capstone Project work.
Total Marks
100 Marks
Theory Weightage
50 Marks (110h)
Practical / Lab
50 Marks (100h)
Total Hours
210 Hours
Class 12 AI Curriculum Blueprint (2026–2027)
The curriculum is structured into three integrated parts: Part A (Employability Skills), Part B (Subject Specific Skills), and Part C (Practical & Capstone Project Work).
Employability Skills
5 advanced units: Communication Skills-IV, Self-Management Skills-IV, ICT Skills-IV (Advanced Spreadsheets), Entrepreneurial Skills-IV, and Green Skills-IV.
Subject Specific Skills
8 in-depth technical units covering Data Science Methodology, Computer Vision, Orange Data Mining, Big Data Analytics, Neural Networks, Generative AI, and Data Storytelling.
Practical & Capstone Work
50-mark hands-on evaluation: Capstone Project (15M) + Documentation (6M) + 3-min Video (4M), Practical File (10M), Lab Exam (10M), and Viva Voce (5M).
Subject Specific Skills – 8 Chapter Modules
Comprehensive study guides containing theory notes, real-world industry case studies, code examples in Python & Orange, and official CBSE Teacher Handbook MCQs with complete answer keys.
Python Programming - II
NumPy array rank & operations, Pandas Series & DataFrames, row and column addition/deletion (.loc[], .drop()), CSV import/export (read_csv, to_csv), handling missing values (isnull, dropna, fillna), and Linear Regression using Scikit-Learn on USA Housing data.
Core Syllabus Highlights:
- NumPy array creation, rank, mean, and standard deviation calculations
- Pandas Series & DataFrames from lists, dictionaries, and NumPy arrays
- Row/Column addition and deletion (.loc[], .drop(axis=0/1), .head(), .tail())
- Missing value treatment (isnull, dropna, fillna with mean/median/zeros)
- Linear Regression algorithm with Scikit-learn (80:20 train-test split)
Data Science Methodology: An Analytic Approach to Capstone Project
John B. Rollins (IBM Analytics) 10-step iterative methodology across 5 modules: Problem to Approach (5W1H & 4 analytics types), Requirements to Collection, Understanding to Preparation & Feature Engineering, Modelling to Evaluation, and Deployment to Feedback. Model validation (Train-Test Split vs K-Fold CV) and evaluation metrics (Confusion Matrix, Precision, Recall, F1, Accuracy, MAE, MSE, RMSE).
Core Syllabus Highlights:
- 10 iterative steps of Data Science Methodology across 5 modules
- 4 Types of Analytics: Descriptive, Diagnostic, Predictive, Prescriptive
- Feature engineering: deriving new features to optimize ML models
- Model Validation: Train-Test Split (80:20, 70:30) vs K-Fold Cross-Validation
- Evaluation Metrics: Confusion Matrix, Precision, Recall, F1-score, MSE, RMSE
Making Machines See
Foundations of Computer Vision (pixels, 0-255 grayscale, 8-bit RGB channels). 5 stages of the CV process: Acquisition, Preprocessing (noise reduction, normalization, resizing 224x224, histogram equalization), Feature Extraction (edge, corner, texture, color), Detection/Segmentation (classification, localization, object detection with R-CNN/YOLO/SSD, semantic vs instance segmentation), and High-Level Processing. Binary Art, Teachable Machine, and Python OpenCV.
Core Syllabus Highlights:
- Image representation in pixels, grayscale (0-255) and RGB color models (16M colors)
- 5 Stages of CV: Acquisition, Preprocessing, Feature Extraction, Detection, High-Level
- Object Detection (R-CNN, YOLO, SSD) vs Image Segmentation (Semantic vs Instance)
- Real-world applications in healthcare, surveillance, self-driving cars, OCR, and 3D modeling
- OpenCV in Python: cv2.imread, cv2.imshow, cv2.resize, and cv2.cvtColor
AI with Orange Data Mining Tool
Component-based visual programming in Orange. Master the Blank Canvas, Widgets, and Connectors across Data, Transform, Visualize, Model, Evaluate, and Unsupervised catalogs. Hands-on applications in Data Science (Iris classification & confusion matrix), Computer Vision (Image Analytics add-on, embeddings, hierarchical clustering dendrograms), and NLP (Text add-on, corpus preprocessing, word clouds).
Core Syllabus Highlights:
- Visual programming interface: Blank Canvas, interactive Widgets, and Connectors
- Default Widget Catalog: Data, Transform, Visualize, Model, Evaluate, Unsupervised
- Data Science: Iris dataset scatter plot, Decision Tree, Test & Score (10-fold CV)
- Computer Vision: Image Analytics add-on, deep learning embeddings & hierarchical clustering
- NLP: Text add-on, Corpus Viewer, text preprocessing, and Word Cloud generation
Introduction to Big Data and Data Analytics
Small Data vs Big Data (328.77M TB daily). Structured, Semi-Structured, and Unstructured data. Advantages, challenges, privacy laws (GDPR, DPDP Act 2023). The 6Vs framework: Volume, Velocity, Variety, Veracity, Value, and Variability with the OnDemandDrama OTT case study. 4 analytics types, 4 global drivers, working stages (Gather, Process, Clean, Analyze), and Mining Data Streams.
Core Syllabus Highlights:
- Small Data vs Big Data: transactional, machine sensor, and social media sources
- Data Types: Structured (RDBMS), Semi-Structured (JSON/XML/CSV), Unstructured (video/audio)
- The 6Vs of Big Data: Volume, Velocity, Variety, Veracity, Value, and Variability
- Working Process: Gather, Process (Batch vs Stream), Clean (Impute), and Analyze
- Mining Data Streams: continuous real-time data flows and future trends
Understanding Neural Networks
Biological neuron inspiration to Artificial Neural Networks (ANN). Input, Hidden, and Output layers; definition of Deep Learning (2+ hidden layers). 7 key components: Neurons, Weights, Activation Functions (Sigmoid, Tanh, ReLU), Bias, Connections, Learning Rules, and Propagation Functions (Forward & Backpropagation via gradient descent). Mathematical working formula w*x+b, surfing & college decision models, and 5 architectures: Perceptron, FFNN, CNN, RNN, and GAN.
Core Syllabus Highlights:
- Biological neuron anatomy translated into Artificial Neural Network layers
- 7 Key Components: Neurons, Weights, Activation Functions, Bias, Connections, Learning Rules, Propagation
- Working of a Neuron: Summation formula (Sigma wi*xi + bias) and step thresholding
- 5 NN Types: Perceptron (1958), Feedforward (FFNN/MLP), CNN, RNN, and GAN
- Hands-on: TensorFlow/Keras Celsius-to-Fahrenheit model & TensorFlow Playground
Generative AI
Principles of Generative AI (Weak AI) creating novel content resembling training samples. Deep learning models: GANs (Generator vs Discriminator) and VAEs (Encoder, Latent Space, Decoder). Comprehensive Generative vs Discriminative AI comparison. Multimodal generation: Image (Canva, DALL-E), Text (ChatGPT, Gemini), Video (Lumiere, deepfakes), Audio (MusicLM, Voicebox). Large Language Models (LLMs) & Transformers (GPT-4o, Gemini 1.5 Pro, LLaMA 3.1, Claude 3.5), limitations, and ethical guidelines (APA citation).
Core Syllabus Highlights:
- What is Generative AI? Creating synthetic images, text, audio, and video
- Working Mechanisms: GANs (Generator vs Discriminator) and VAEs (Encoder, Latent Space, Decoder)
- Generative vs Discriminative Models: comparison matrix and algorithms
- Large Language Models (LLMs): Transformer architecture, top models, and LLaMA case study
- Ethical & Social Concerns: Deepfakes, algorithmic bias (HireVue, Clearview AI), and APA citation
Data Storytelling
Evolution from cave paintings to digital data storytelling. Blending the 3 essential elements: Data + Narrative + Visuals to inform, persuade, and inspire. Narrative arc using Freytag's Pyramid: Exposition/Setting, Rising Action/Insights, Climax/Aha Moment, Falling Action/Solution, and Resolution/Outcome. Visualization selection matrix (Word Cloud, FacetGrid, Line, Bar, Pie, Scatter, Candlestick, Map charts). Case studies: Mid-Day Meal Scheme (MDMS) dropout rates and COVID-19 vaccination impact.
Core Syllabus Highlights:
- The Power of Stories: Memorability, persuasion, engagement, and actionable insights
- 3 Essential Elements of Data Storytelling: Data (foundation), Narrative (storyline), Visuals (scene)
- Narrative Structure: Adapting Freytag's Pyramid to craft insightful data stories
- Visualizations for Different Data: Matching text, numeric, temporal, stock, and geographic data
- Real-world Case Studies: Mid-Day Meal Scheme (MDMS r = -0.9313) and COVID-19 vaccine trends
Practical Examination & Capstone Project Blueprint
The 50-mark internal & external lab evaluation criteria prescribed in the official CBSE Class 12 curriculum for academic session 2026–2027.
AI Model & Prototype
Team execution (3–5 students) of an end-to-end AI project aligned with United Nations SDGs using Data Science, Computer Vision, or NLP.
Project Report
Comprehensive project documentation covering Problem Statement (5W1H), SDG rationale, Data requirements, algorithms, and evaluation metrics.
3-Minute Video Presentation
Precise 3-minute recorded video presentation showcasing the problem, SDG impact, live demo of the working prototype, and conclusion.
Practical File (10 Activities)
Prescribed file containing minimum 6 Python programs (Pandas & Scikit-Learn), 3 Orange workflows, and 1 complete Data Story problem.
Lab Test (Python & Orange)
Hands-on practical coding examination covering Python data manipulation / regression and visual model building in Orange Data Mining.
Viva Voce
Oral questioning by external & internal examiners on the Capstone Project, Data Science Methodology, Neural Networks, and Practical File.
Prescribed Hardware & Software Lab Requirements
Class 12 AI High-Yield Recall Points
Frequently Asked Questions (FAQ)
Master Class 12 Artificial Intelligence with 1:1 Live Mentorship
Get 1-on-1 personalized guidance from certified AI educators. Master Python machine learning pipelines, Orange visual data mining, Neural Networks, Generative AI, Capstone project mentoring, and score 100/100 in CBSE Class 12 AI exams.