Sample Question Papers (SQPs) play a crucial role in effective exam preparation, especially for subjects like Artificial Intelligence for CBSE Class 12 where both conceptual understanding and application skills are important. Practicing SQPs helps students become familiar with the latest CBSE exam pattern, question types, and marking scheme. It also improves time management, allowing students to plan how much time to allocate to each section during the actual exam. Most importantly, solving these papers helps identify weak areas, enabling focused revision and boosting overall confidence before the board exams.
CBSE Class 12 Artificial Intelligence (Code 843) Sample Question Paper for 2025-26 with complete solutions is designed as per the latest syllabus and guidelines. This paper covers all major topics and includes a balanced mix of MCQs, case-based, and subjective questions to simulate the real exam experience. The detailed solutions provided help students understand the correct approach, improve answer-writing skills, and avoid common mistakes. By practicing this paper seriously, students can strengthen their preparation, gain clarity in concepts, and enhance their chances of scoring high marks in the board examination.
CBSE Class 12 AI 843 Sample QP 2025-26 with Solution
You can download the free PDF of this sample pre-board question paper from our website and start practicing today. Download Link available in last.
The Year-Wise CBSE Class 12 Artificial Intelligence (843) Pre-Board and Sample Question Papers with Solutions are provided at the bottom of this post.
CBSE Class 12 AI 843 Sample Question Paper Session (2025-26)
CBSE | DEPARTMENT OF SKILL EDUCATION
ARTIFICIAL INTELLIGENCE (SUBJECT CODE - 843)
Sample Question Paper (SQP) Solution for Class XII (Session 2025-2026)
Max. Time: 2 Hours
Max. Marks: 50
General Instructions:
- Please read the instructions carefully.
- This Question Paper consists of 21 questions in two sections: Section A & Section B.
- Section A has Objective type questions whereas Section B contains Subjective type questions.
- Out of the given (5 + 16 =) 21 questions, a candidate has to answer (5 + 10 =) 15 questions in the allotted (maximum) time of 2 hours.
- All questions of a particular section must be attempted in the correct order.
- SECTION A - OBJECTIVE TYPE QUESTIONS (24 MARKS):
- This section has 05 questions.
- Marks allotted are mentioned against each question/part.
- There is no negative marking.
- Do as per the instructions given.
- SECTION B – SUBJECTIVE TYPE QUESTIONS (26 MARKS):
- This section has 16 questions.
- A candidate has to do 10 questions.
- Do as per the instructions given.
- Marks allotted are mentioned against each question/part.
SECTION A - OBJECTIVE TYPE QUESTIONS
Q.1 Answer any 4 out of the given 6 questions on Employability Skills. (1X4=4 Marks)
i. During a group discussion, Priya is explaining her point of view on a project idea. Midway through her explanation, Arun interrupts and starts sharing his own thoughts without letting Priya complete her sentence.
If you were in Arun’s place and wanted to follow the T of RESPECT in active listening, what should you do?
a) Interrupt politely if you feel your point is important
b) Start speaking as soon as you think you understand her point
c) Speak at the same time so both ideas are heard together
d) Wait until Priya finishes speaking, then respond to her points
ii. ________is a trait, wherein, individuals show tendency towards anxiety, self-doubt, depression, shyness and other similar negative feelings. (Fill in the blank)
Ans: Neuroticism
iii. Match the following skills with their correct descriptions:
| Skills | Description |
|---|---|
| 1. Positive thinking | a. To be aware of your strengths and make the best out of them. |
| 2. Result orientation | b. To think that one can get things done and be happy. |
| 3. Self-awareness | c. To dream big and achieve the desired or set goals. |
a) 1 → b, 2 → c, 3 → a
b) 1 → a, 2 → b, 3 → c
c) 1 → b, 2 → a, 3 → c
d) 1 → c, 2 → a, 3 → b
iv. What happens when you change the layout of a slide?
a) The arrangement of content (text, images, shapes) changes
b) The format of the text changes
c) New slide is inserted
d) The title gets aligned to the center of the slide
v. Assertion (A): Economists, psychologists, sociologists, and management experts define entrepreneurs differently.
Reason (R): The definition of an entrepreneur varies depending on whether the focus is on resources, psychological drive, societal contribution, or vision with action.
a) Both A and R are correct, and R is the correct explanation of A
b) Both A and R are correct, but R is not the correct explanation of A
c) A is correct, but R is incorrect
d) A is incorrect, but R is correct
vi. Appropriate technology is small-scale, environment-friendly technology suited to local needs. It gets the job done using locally available resources in a sustainable manner. (State whether this is True/False)
Ans: True
Q.2 Answer any 5 out of the given 6 questions. (1X5=5 Marks)
i. Match the correct question from Column A to the correct step of the Data Science Methodology in Column B.
| Column A | Column B |
|---|---|
| 1. In what way can the data be visualized to get to the required answer? | a. Data Understanding |
| 2. How can you use the data to answer the question? | b. Business understanding |
| 3. What is the problem that you are trying to solve? | c. Analytic approach |
| 4. Is the data collected representative of the problem to be solved? | d. AI Modelling |
a) 1 → a, 2 → b, 3 → c, 4 → d
b) 1 → d, 2 → c, 3 → b, 4 → a
c) 1 → b, 2 → a, 3 → d, 4 → c
d) 1 → c, 2 → b, 3 → a, 4 → d
ii. The given diagram shows the working of a vision-based system that identifies objects in an image. What is the other name for this technology?
Ans: Machine Vision
iii. Which law suggests that increasing computing power over time makes it easier to handle large volumes of data?
a) Boyle's Law
b) Murphy's Law
c) Newton's Law
d) Moore's Law
iv. An Artificial Neural Network (ANN) with two or more hidden layers is known as a _______ (Fill in the blank)
Ans: Deep Neural Network
v. What will come in place of ‘1’ and ‘2’?
Ans: 1 → user prompt; 2 → training data
vi. In the context of data storytelling, a valuable and meaningful observation or understanding derived from analysing data is called:
a) Prediction
b) Pattern
c) Insight
d) Correlation
Q.3 Answer any 5 out of the given 6 questions. (1X5=5 Marks)
i. _______ is the process of selecting, modifying, or creating new features from raw data to improve the performance of machine learning models. (Fill in the blank)
Ans: Feature Engineering
ii. What will come in place of 1 and 2?
a) True Positive, False Positive
b) False Positive, True Negative
c) True Negative, False Negative
d) False Negative, False Positive
iii. Match the image processing technique in Column A with its correct description in Column B.
| Column A | Column B |
|---|---|
| 1. Edge Detection | a. Finds points where two or more edges meet |
| 2. Corner Detection | b. Identifies boundaries between regions with large intensity change |
| 3. Texture Analysis | c. Extracts features like smoothness or repetition |
| 4. Colour-based Feature Extraction | d. Measures colour distribution to distinguish objects |
a) 1 → a, 2 → b, 3 → c, 4 → d
b) 1 → a, 2 → c, 3 → b, 4 → d
c) 1 → b, 2 → a, 3 → c, 4 → d
d) 1 → c, 2 → b, 3 → a, 4 → d
iv. Big Data refers to extremely large and complex datasets that regular computer programs and databases cannot handle. Which of the following is NOT one of its main sources?
a) Printed Library catalogue data
b) Transactional data
c) Machine data
d) Social data
v. Identify the odd one out from the following activation functions:
a) Sigmoid Function
b) Tanh Function
c) ReLU (Rectified Linear Unit)
d) Linear Regression
vi. Refer to the Venn diagram that illustrates the three key elements of Data Storytelling: Narrative, Visuals, and Data.
What term should replace the question mark (?) in the center, where all three elements intersect?
Ans: Change
Q.4 Answer any 5 out of the given 6 questions. (1X5=5 Marks)
i. Assertion(A): The diagnostic measures phase is used to ensure that the model is working as intended and to refine it if necessary.
Reason (R): In this phase, predictive models can be evaluated using decision trees, while descriptive models can be tested with known outcomes to check alignment with the design.
a) Both A and R are correct, and R is the correct explanation of A
b) Both A and R are correct, but R is not the correct explanation of A
c) A is correct, but R is incorrect
d) A is incorrect, but R is correct
ii. When a computer processes an image, it perceives it as a collection of tiny squares known as ______.
Ans: Pixels
iii. Which emerging technology in Big Data analytics promises unprecedented processing power and can solve complex problems much faster than classical computers?
a) Cloud Computing
b) Quantum Computing
c) Edge Computing
d) Parallel Processing
iv. In a neural network, which component defines how weights and biases are adjusted during training?
a) Activation function
b) Learning rule
c) Loss function
d) Hidden layer
v. Name the two neural networks used in a Generative Adversarial Network (GAN).
Ans: Generator, Discriminator
vi. Match the type of Generative AI application to the correct AI tool.
| AI Tool | Application Type |
|---|---|
| 1. Stable Diffusion | a. Audio Generation |
| 2. ChatGPT | b. Text Generation |
| 3. Google’s Lumiere | c. Video Generation |
| 4. Meta’s AI Voicebox | d. Image Generation |
a) 1 → c, 2 → d, 3 → a, 4 → b
b) 1 → a, 2 → c, 3 → b, 4 → d
c) 1 → d, 2 → b, 3 → c, 4 → a
d) 1 → c, 2 → b, 3 → d, 4 → a
Q.5 Answer any 5 out of the given 6 questions. (1X5=5 Marks)
i. Feedback is the last stage of data science methodology that helps to refine the model and assess it for performance and impact. (State whether this is True/False)
Ans: True
ii. Assertion (A): Image normalization adjusts the pixel values of an image so they fall within a consistent range, such as 0–1 or -1 to 1.
Reason (R): It ensures that all images in a dataset have a similar scale, which helps the model learn better.
a) Both A and R are correct, and R is the correct explanation of A
b) Both A and R are correct, but R is not the correct explanation of A
c) A is correct, but R is incorrect
d) A is incorrect, but R is correct
iii. The image shows bounding boxes and labels for multiple objects like a bicycle, dog, and car.
Which computer vision task does this represent?
a) Image Classification
b) Object Detection
c) Instance Segmentation
d) Semantic Segmentation
iv. Match the type of neural network in Column A with its corresponding task in Column B.
| Column A | Column B |
|---|---|
| 1. Perceptron | a. Classification |
| 2. Convolutional Neural Network (CNN) | b. Object detection |
| 3. Recurrent Neural Network (RNN) | c. Machine translation |
a) 1 → c, 2 → a , 3 → b
b) 1 → a, 2 → b, 3 → c
c) 1 → b, 2 → c, 3 → a
d) 1 → c, 2 → b, 3 → a
v. ______ is the compressed representation of the actual data in Variational Autoencoders. (Fill in the blank)
Ans: Latent space
vi. A chart used in stock, forex, commodity, and option trading that shows price movements (open, high, low, close) is a ________.
a) Line chart
b) Bar chart
c) Facet grid chart
d) Candle stick chart
SECTION B - SUBJECT TYPE QUESTIONS
Answer any 3 out of the given 5 questions on Employability Skills. (2x3=6 Marks)
Answer each question in 20-30 words.
Q.6 Wow! Oh no! Thanks! Help!
What are these words known as? Explain.
Ans: The words - “Wow!”, “Oh no!”, “Thanks!”, “Help!” - are known as interjections.
They are words that express strong emotions, such as happiness, surprise, anger or pain.
Q.7 Write any two steps a person can take to overcome personality disorders.
Ans: (i) Talk to someone. Most often, it helps to share your feelings.
(ii) Build confidence in your ability to handle difficult situations.
(iii) Stay positive by choosing words like ‘challenges’ instead of ‘problems’.
Q.8 In a spreadsheet, you can position the text in a cell to the left, right, or center.
a) Name this feature of a spreadsheet.
Ans: Alignment
b) How are numbers and text positioned in a cell by default?
Ans: Numbers are right-aligned by default.
Text is left-aligned by default.
Q.9 Identify the type of attitude the following entrepreneur is showing:
a) Ganesh was always interested in hairstyling. He knew the latest international styles and would keep trying them on his friends. But his family wanted him to work in a software company. Encouraged by friends, he spoke to his family, and convinced them of his talent and business skills. With their support, he set up his own salon.
Ans: Taking initiative
b) Malvika has a book selling business. One day, a shipment of her books gets lost. This creates a lot of problems for her customers. She apologises to them and works hard for two days to get a new shipment by the next day.
Ans: Perseverance
Q.10 What is meant by waste exchange? How does waste exchange help in reducing waste disposal?
Ans: Waste exchange is when the waste product of one process becomes the raw material for another. It helps in reducing waste disposal by reusing waste, thereby minimizing the amount sent to landfills.
Answer any 4 out of the given 6 questions in 20-30 words each. (2x4=8 Marks)
Q.11 Define Data Science Methodology. How many steps are there in it?
Ans: Data Science Methodology is a structured process that provides a systematic and iterative approach to solving problems using data. It helps data scientists move from problem understanding to solution deployment effectively.
It consists of 10 steps.
(i) Business Understanding (ii) Analytic Approach (iii) Data Requirements (iv) Data Collection (v) Data Understanding (vi) Data Preparation (vii) Modeling (viii) Evaluation (ix) Deployment (x) Feedback
Q.12 Computer vision, a vital part of artificial intelligence, faces several hurdles as it strives to make sense of the visual world around us. Explain any two challenges related to computer vision.
Ans: Two challenges of Computer Vision are:
(i) Difficulty in Image Acquisition: Capturing high-quality images is challenging due to variations in lighting, angle, scale, and occlusion. These factors make it difficult for computer vision systems to correctly interpret visual data.
(ii) Reasoning and Analytical Issues: Computer vision not only identifies objects but also needs to interpret them correctly. Lack of strong reasoning and analytical ability makes it difficult to extract meaningful insights from images.
Q.13 Differentiate between Batch Processing and Stream Processing.
Ans:
| Batch Processing | Stream Processing |
|---|---|
| Processes large volumes of data all at once within a specific time span. | Takes more time to process data. |
| Processes continuous streams of data immediately as it is produced. | Takes less time (seconds or milliseconds) to process data. |
Q.14 Consider the following perceptron that predicts the academic performance of a student with inputs, weights, and a bias of –2. Calculate Output/Predicted outcome (ŷ) for the given scenario.
| Factor | Input | Weight |
|---|---|---|
| Study Hours | 1 | 6 |
| Distractions (TV/Games) | 0 | 3 |
| Sleep Quality | 1 | 2 |
Ans: ŷ = ∑wixi+bias
= w1x1 + w2x2 + w3x3 + bias
Substituting values of w1, x1 , w2, x2 ,w3 , x3, bias, we get
ŷ = (6 X 1) + (3 X 0) + (2 X 1) – 2
= 6 + 0 + 2 − 2
= 6
Q.15 What distinguishes generative models from discriminative models?
Ans: Generative models learn the underlying data distribution and can generate new data samples similar to the original data.
Discriminative models focus on distinguishing between different classes by learning decision boundaries.
Q.16 List the steps involved in creating a compelling story through data.
Ans: • Collect and organize the data.
• Analyze the data to identify patterns and relationships.
• Use appropriate visualization tools to present the data.
• Create a clear and simple narrative for the audience.
Answer any 3 out of the given 5 questions in 50-80 words each. (4x3=12 Marks)
Q.17 Validating the machine learning model during the training and development stages is crucial for ensuring accurate predictions. Differentiate between the Train–Test Split and Cross-Validation methods, and support your answer with a neat, labelled diagram.
Ans:
| Train-Test Split | Cross Validation |
|---|---|
 |
 |
| Normally applied on large data sets. | Normally applied on small data sets. |
| Divides the data into training data set and testing data set. | Divides a dataset into subsets (folds), trains the model on some folds, and evaluates its performance on the remaining data. |
| Clear demarcation on training data and testing data. | Every data point at some stage could be in either testing or training data set. |
Q.18 A global e-commerce company handles millions of transactions every day. Identify the characteristics of Big Data you can relate to in the following scenarios and explain them in detail:
a) Customers place orders online, and the website records thousands of clicks, searches, and transactions every second.
Ans: Velocity refers to the speed at which data is generated, processed, and analyzed.
In today’s digital world, massive amounts of data are produced every second.
Example: Thousands of clicks, searches, and transactions happening every second show high-speed data generation and often require real-time processing.
b) The company’s servers store petabytes of customer orders, payment details, and product listings collected over years.
Ans: Volume refers to the huge amount of data generated and stored.
This data can range from terabytes to petabytes or even exabytes.
Example: Storing petabytes of customer orders, payment details, and product listings over time represents large data volume.
c) The stored data comes in different formats — structured (databases of products), semi-structured (XML, JSON order files), and unstructured (customer reviews, product photos, videos).
Ans: Variety refers to different types and formats of data.
Data can be structured (tables), semi-structured (XML/JSON), or unstructured (text, images, videos).
Example: Product databases, XML/JSON files, and customer reviews/photos/videos show data diversity.
d) Sometimes, the system collects incomplete or duplicate records, and before analysis, the team removes errors to ensure the information is accurate and trustworthy.
Ans: Veracity refers to the quality, accuracy, and reliability of data.
Data may contain errors, duplicates, or inconsistencies.
Example: Cleaning incomplete or duplicate records ensures trustworthy and accurate data.
Q.19 Explain the role of neural networks in the future of AI.
Ans: How Neural Networks are Helping AI Grow
• Advancing Deep Learning: Neural Networks with multiple layers power Deep Learning, helping
machines understand complex patterns in large datasets.
• Improving Accuracy and Efficiency: They analyze vast data and make highly accurate predictions (e.g., detecting diseases from medical images), boosting efficiency in real-world tasks.
• Supporting Autonomous Systems: Neural networks provide the perception and decision-making capabilities needed for autonomous systems (e.g., self-driving cars, robots, drones), enabling real-time sensing, control, and safe automation.
• Personalizing Experiences: By analyzing user preferences, Neural Networks power recommendation systems (shopping, music, movies), creating tailored experiences.
Q.20 Explain how Meta’s LLaMA is unique compared to traditional LLMs with a neat diagram.
Ans: LLaMA is unique compared to traditional Large Language Models (LLMs) in the following ways:
1. Efficient and Open Training Approach:
LLaMA is trained primarily on publicly available data (text and code), unlike many traditional LLMs that rely heavily on proprietary datasets. This promotes transparency and research accessibility.
It also uses efficient training techniques, allowing strong performance with less computational power, making it more scalable.
2. Multiple Model Sizes (Flexibility):
Meta released LLaMA in different sizes (e.g., 7B, 13B, 33B, 65B parameters).
- Smaller models → suitable for devices with limited resources
- Larger models → better for complex NLP tasks
This makes LLaMA highly flexible and practical.
3. High Performance with Less Data/Compute:
Despite using public datasets and fewer resources, LLaMA achieves performance comparable to or better than larger proprietary models in tasks like:
- Text summarization
- Question answering
- Language understanding
This shows that better training strategy can outperform sheer model size.
Q.21 The image illustrates Freytag’s Pyramid, a classic narrative structure.
While (1) Exposition introduces the story and characters, identify and briefly explain the stages labeled as 2, 3, 4, and 5 in the diagram.
Ans: 2. Rising action: The series of events that build up to the climax of the story.
3. Climax: The most intense or important point within the story. It is often an event in which the fortune of the protagonist turns for the better or worse in the story.
4. Falling action: The rest of the events that unravel after the main conflict has occurred, but before the final outcome is decided.
5. Conclusion: The conclusion of the story where all of the conflicts are resolved and outstanding details are explained.
Download Now - Class 12 AI 843 Sample Question Paper 2025-26 with Solution
