CBSE Class 12 Artificial Intelligence · Practical viva
CBSE Class 12 Artificial Intelligence Practical Viva Questions, with Answers
The viva carries marks in the CBSE Class 12 Artificial Intelligence practical, and it is the part students prepare least. Below are 16 questions examiners actually ask, grouped by topic, each with a short answer you can say out loud.
AI, ML and the project cycle
1. What is the difference between AI, machine learning and deep learning?
Answer: AI is any machine behaving intelligently. Machine learning is AI that learns from data rather than fixed rules. Deep learning is machine learning using neural networks with many layers.
2. Name the stages of the AI Project Cycle.
Answer: Problem Scoping, Data Acquisition, Data Exploration, Modelling and Evaluation.
3. What are the 4Ws of problem scoping?
Answer: Who has the problem, What the problem is, Where it happens, and Why solving it matters.
4. What is supervised learning?
Answer: Learning from data that is already labelled, so the model is told the right answer while training. Unsupervised learning finds patterns in unlabelled data.
5. What is the difference between classification and regression?
Answer: Classification predicts a category. Regression predicts a number.
Data and modelling
1. Why is data split into training and testing sets?
Answer: So the model can be judged on data it has never seen. Testing on the training data would only show that it memorised.
2. What is overfitting?
Answer: When a model learns the training data too closely, including its noise, so it does well in training and badly on new data.
3. Why must data be cleaned before modelling?
Answer: Missing values, duplicates and errors are learned as if they were real patterns. Bad data gives a confident wrong model.
4. What is normalisation?
Answer: Rescaling values into a common range, usually 0 to 1, so one large-valued feature does not dominate the others.
Evaluation
1. What is a confusion matrix?
Answer: A table of True Positives, True Negatives, False Positives and False Negatives, comparing what the model predicted against what was actually true.
2. What is the difference between precision and recall?
Answer: Precision asks: of everything the model flagged, how much was right. Recall asks: of everything that was actually positive, how much did it find.
3. Why is accuracy a poor measure on imbalanced data?
Answer: If 99 of 100 cases are negative, a model that always says negative is 99% accurate and completely useless. Precision, recall and F1 expose that.
4. What is the F1 score?
Answer: The harmonic mean of precision and recall, used when both matter and the classes are imbalanced.
Ethics
1. What is AI bias, with an example?
Answer: When a model's output systematically disadvantages a group, because the data it learned from did. A hiring model trained on past hires can repeat past bias.
2. Can an AI system be held responsible for a mistake?
Answer: No. Accountability stays with the people and organisations that build and deploy it.
3. What is data privacy in an AI project?
Answer: Collecting only the data needed, with consent, storing it safely, and not using it for a purpose the person did not agree to.
Want the full set?
More questions, the practical file format and the topics examiners come back to — sent on WhatsApp.
Get it on WhatsApp Book a free demo classPlease read: these are practice questions written by Kwickprep to help you revise. They are not past examination papers and not a CBSE publication. Questions asked in your school’s viva or examination may differ, and the syllabus can change between sessions — the board’s own document at cbseacademic.nic.in is always the final word. Tell us if you spot a mistake and we will correct it.