Class 9 AI(417) – AI Reflection Project Cycle and Ethics MCQs
AI Reflection Project Cycle and Ethics MCQs built strictly on the updated CBSE Class 9 AI syllabus, covering every key topic from the chapter. Strengthen your concepts, test your understanding, and prepare with exam-focused questions designed to challenge you—not just what’s easy.
So, why wait? 🚀 Let’s dive in and start mastering the most important CBSE Class 9 AI MCQs! 📚✨
Q1. A machine collects data, analyses it, learns from it and improves its performance without being explicitly instructed for every situation. Which concept best describes this capability?
a) Manual computing
b) Artificial Intelligence
c) Data entry
d) Remote control
Q2. Which statement best represents the core idea of Artificial Intelligence?
a) Storing large amounts of data
b) Making computers perform only arithmetic operations
c) Enabling machines to perform tasks requiring human-like intelligence
d) Replacing all human activities with machines
Q3. Which of the following is NOT listed as a trait of an artificially intelligent machine?
a) Mimics human intelligence
b) Solves real-world problems
c) Improves on its own from past experiences
d) Requires manual reprogramming after every task
Q4. A smartphone unlocks itself after comparing the user’s face with previously captured facial features. Which AI domain is primarily involved?
a) NLP
b) Statistical Data
c) Computer Vision
d) Data Acquisition
Q5. Siri or Alexa listens to a spoken request, identifies patterns in speech and provides a response. This primarily demonstrates:
a) Computer Vision
b) Natural Language Processing
c) Statistical Data
d) System Mapping
Q6. Which type of data is primarily associated with Natural Language Processing?
a) Images
b) Videos
c) Textual data
d) Numerical tables only
Q7. Which AI domain works mainly with images and videos?
a) NLP
b) Computer Vision
c) Statistical Data
d) Problem Scoping
Q8. Statistical Data is associated mainly with:
a) Analysing numerical/tabular data
b) Understanding spoken language
c) Processing images
d) Creating system maps
Q9. A banking organisation analyses customer profiles, previous expenditures and other variables to estimate the probability of loan default. Which area is most directly represented?
a) Fraud and risk detection
b) Computer Vision
c) NLP
d) Deployment
Q10. Which of the following is the best example of Computer Vision?
a) Autocorrecting a sentence
b) Recognising a face
c) Predicting loan risk from numerical data
d) Generating a text response
Q11. Semantris is an AI game based on which domain?
a) Computer Vision
b) NLP
c) Statistical Data
d) Deployment
Q12. Medical imaging applications that analyse scan images primarily demonstrate:
a) Computer Vision
b) NLP
c) Problem Scoping
d) Rule-Based AI only
Q13. A system that converts 2D medical scan images into interactive 3D models is most closely associated with:
a) Computer Vision
b) NLP
c) Data Acquisition
d) Ethics
Q14. Which AI domain would be most suitable for detecting whether people are wearing masks from camera images?
a) NLP
b) Computer Vision
c) Statistical Data
d) System Mapping
Q15. Text editors and autocorrect systems are most closely associated with:
a) NLP
b) Computer Vision
c) Deployment
d) Data Acquisition
Q16. Which is the correct first stage of the AI Project Cycle?
a) Data Acquisition
b) Problem Scoping
c) Modelling
d) Evaluation
Q17. Which stage follows Problem Scoping?
a) Deployment
b) Evaluation
c) Data Acquisition
d) Modelling
Q18. The correct sequence of the AI Project Cycle is:
a) Problem Scoping → Data Acquisition → Data Exploration → Modelling → Evaluation → Deployment
b) Data Acquisition → Problem Scoping → Modelling → Evaluation → Deployment → Data Exploration
c) Problem Scoping → Modelling → Data Acquisition → Evaluation → Data Exploration → Deployment
d) Data Exploration → Problem Scoping → Data Acquisition → Modelling → Deployment → Evaluation
Q19. The main purpose of Problem Scoping is to:
a) Train the final model
b) Identify a problem and set a clear goal
c) Deploy the model
d) Calculate accuracy
Q20. The AI Project Cycle helps us to:
a) Create AI projects slower but with more effort
b) Create better AI projects easily and faster
c) Avoid using data
d) Skip the evaluation stage
Q21. Which two benefits are associated with using the AI Project Cycle?
a) Complexity and Redundancy
b) Efficiency and Modularity
c) Ambiguity and Duplication
d) Cost and Time only
Q22. Identifying a problem and having a vision to solve it is called:
a) Data Acquisition
b) Problem Scoping
c) Evaluation
d) Deployment
Q23. Problem Scoping involves:
a) Identifying a problem and having a vision to solve it
b) Only collecting data
c) Only writing code
d) Only testing a model
Q24. A broad area such as Agriculture is called a:
a) Data feature
b) Theme
c) Model
d) Stakeholder
Q25. In the 4Ws Problem Canvas, “Who?” primarily identifies:
a) The location of the problem
b) The stakeholders
c) The evidence
d) The benefit
Q26. Which 4W focuses on the nature of the problem and evidence proving that it exists?
a) Who
b) What
c) Where
d) Why
Q27. Newspaper articles and media reports can be used under which 4W?
a) Who
b) What
c) Where
d) Why
Q28. The “Where?” block examines:
a) Stakeholders
b) Evidence
c) Context, situation and location
d) Benefits of the solution
Q29. The “Why?” block mainly focuses on:
a) The people affected
b) The evidence
c) Location
d) Benefits of solving the problem
Q30. A person directly affected by a problem and expected to benefit from its solution is a:
a) Stakeholder
b) Data feature
c) Model
d) Algorithm
Q31. The Problem Statement Template is used to:
a) Summarise key points of the problem in one place
b) Train an AI model
c) Calculate accuracy
d) Deploy an application
Q32. In the Problem Statement Template, “[stakeholders]” corresponds to:
a) Who
b) What
c) Where
d) Why
Q33. “Has a problem that [issue, problem, need]” corresponds to:
a) Who
b) What
c) Where
d) Why
Q34. “An ideal solution would [benefit of solution]” corresponds to:
a) Who
b) What
c) Where
d) Why
Q35. Data can be defined as:
a) Only numbers
b) Information, facts and statistics collected for reference or analysis
c) Only images
d) Only text
Q36. Data Acquisition in the AI Project Cycle mainly involves:
a) Testing the AI model
b) Collecting relevant and authentic data for the project
c) Framing the goal of the project
d) Deploying the model on mobile apps
Q37. In AI, the data used to train a machine is called:
a) Testing Data
b) Training Data
c) Deployment Data
d) Evaluation Data
Q38. The dataset used to check how well a trained model performs is called:
a) Training Data
b) Testing Data
c) Acquisition Data
d) Feature Data
Q39. For an AI project to be efficient, the training data should be:
a) Random and unrelated
b) Authentic and relevant to the problem statement
c) As large as possible regardless of relevance
d) Collected only from social media
Q40. Data features refer to:
a) The type of data that needs to be collected
b) The final output
c) The AI model
d) The deployment method
Q41. Which of the following is a data feature for salary prediction?
a) Salary amount
b) Increment percentage
c) Bonus
d) All of the above
Q42. Data obtained from random websites may be unsuitable because:
a) It is always expensive
b) Its accuracy may not be provable
c) It cannot be downloaded
d) It is always private
Q43. Which portal is mentioned in the handbook as an open-source government data source?
a) data.gov.in
b) games.gov.in
c) schooldata.ai
d) private.gov.in
Q44. Extracting private data without appropriate basis can be:
a) An offence
b) A data feature
c) A testing method
d) A system map
Q45. The primary purpose of a System Map is to:
a) Write programming code
b) Understand relationships between elements of a system
c) Store training data
d) Calculate model accuracy
Q46. In a System Map, relationships between elements are represented using:
a) Tables only
b) Arrows
c) Paragraphs
d) Photographs
Q47. A loop in a System Map represents:
a) A chain of causes and effects
b) A programming error
c) A dataset
d) A stakeholder
Q48. A longer arrow in a System Map represents:
a) Greater accuracy
b) A longer time for change to happen
c) More data
d) More stakeholders
Q49. A longer arrow in a System Map generally represents:
a) Data feature
b) Time delay
c) Training loop
d) Feedback error
Q50. If X → Y has a “+” sign and X increases, Y generally:
a) Decreases
b) Also increases
c) Becomes zero
d) Remains unrelated
Q51. If X → Y has a “–” sign and X increases, Y:
a) Increases
b) Decreases
c) Remains unchanged
d) Becomes undefined
Q52. Which system is specifically used as an example of System Mapping?
a) Banking system
b) Water Cycle
c) School timetable
d) Traffic signal
Q53. Data Exploration is carried out mainly to:
a) Discover trends, relationships and patterns
b) Delete all data
c) Deploy the model
d) Replace the problem statement
Q54. Why is data visualisation useful?
a) It helps humans comprehend information quickly
b) It makes data invisible
c) It eliminates the need for data
d) It automatically guarantees accuracy
Q55. Data Exploration can help decide:
a) Which model or models may be suitable
b) Which student should use the computer
c) Which government portal to visit
d) Which ethics principle to ignore
Q56. Graphical representation helps communicate:
a) Trends and patterns
b) Passwords
c) Private information
d) Source code only
Q57. A student plots singer popularity to investigate which singer is more likely to have a song reach the Billboard top 10. This is an example of:
a) Data Visualization for pattern identification
b) Deployment
c) Problem Scoping only
d) Privacy protection
Q58. Which stage occurs immediately before Modelling?
a) Problem Scoping
b) Data Acquisition
c) Data Exploration
d) Deployment
Q59. Which of the following is NOT the purpose of Data Exploration?
a) Understanding patterns
b) Identifying relationships
c) Deciding model strategy
d) Deploying the final solution
Q60. Which website is suggested in the handbook to explore various data visualisation techniques?
a) data.gov.in
b) datavizcatalogue.com
c) moralmachine.net
d) rockpaperscissors.ai
Q61. AI is best described as:
a) A subset of Deep Learning
b) An umbrella term covering ML and DL
c) A type of database
d) A graph
Q62. Artificial Intelligence, Machine Learning and Deep Learning are related such that:
a) DL is the broadest term, AI is the narrowest
b) ML is completely separate from AI
c) AI is the umbrella term, ML is a subset of AI, and DL is a subset of ML
d) AI and DL are unrelated
Q63. Machine Learning enables machines to:
a) Improve at tasks with experience
b) Work without any data
c) Avoid testing
d) Follow only fixed instructions
Q64. Deep Learning is associated with:
a) Vast amounts of data
b) No data
c) Only manually written rules
d) Only text data
Q65. A machine learns from its mistakes and uses that experience in its next execution. This best describes:
a) Machine Learning
b) Data Acquisition
c) Rule-Based static learning
d) Data Visualization
Q66. Which of the following AI models is developed based on rules defined by the developer?
a) Learning-Based Approach
b) Rule-Based Approach
c) Deep Learning Approach
d) Evaluation-Based Approach
Q67. A model that uses huge amounts of data to train itself is most closely associated with:
a) Deep Learning
b) Problem Scoping
c) System Mapping
d) Data Cleaning
Q68. AI Modelling refers to:
a) Developing algorithms/models that can be trained to produce intelligent outputs
b) Collecting newspaper articles
c) Drawing a 4Ws canvas
d) Deploying an app only
Q69. In a Rule-Based approach, rules are generally:
a) Defined by the developer
b) Automatically invented after deployment
c) Removed after training
d) Taken only from testing data
Q70. A Rule-Based model is considered static because:
a) It does not consider changes made to the original training dataset
b) It cannot produce outputs
c) It does not use any rules
d) It always changes itself
Q71. In a Learning-Based approach, the machine:
a) Learns from the data and adapts to changes
b) Only follows manually supplied rules
c) Cannot handle exceptions
d) Never uses testing data
Q72. A dataset contains labelled images of apples and bananas. The model learns features to distinguish between them. This illustrates:
a) Learning-Based Approach
b) Rule-Based Approach
c) Data Acquisition only
d) Deployment
Q73. Which approach is more adaptive to changes in data?
a) Rule-Based
b) Learning-Based
c) Both are equally static
d) Neither
Q74. In a Learning-Based approach, the desired output is provided along with:
a) Data
b) Only rules
c) Deployment software
d) Ethics principles
Q75. The main objective of Evaluation is to:
a) Test different models and choose the best model
b) Select a theme
c) Collect private data
d) Create a problem statement
Q76. Statement I: Privacy requires protection of personal data.
Statement II: AI systems do not need to tell users how their personal data is used.
a) Both are true
b) Both are false
c) I is true but II is false
d) I is false but II is true
Q77. Evaluation determines model reliability by comparing model outputs with:
a) Actual answers
b) Themes
c) Stakeholders
d) System maps
Q78. Why should training data not normally be used for evaluation?
a) The model may simply remember the training data
b) Training data cannot be stored
c) Training data is always private
d) It contains no labels
Q79. When a model performs extremely well on training data because it has effectively memorised it, the handbook calls this:
a) Under-processing
b) Overfitting
c) Deployment
d) Data Acquisition
Q80. Which dataset is separated during Data Acquisition and later used for evaluating the trained model?
a) Testing Data
b) Theme Data
c) Stakeholder Data
d) System Map Data
Q81. ROC in model evaluation is used as a metric to:
a) Store training data
b) Find out the accuracy of a model
c) Scope the problem
d) Acquire data
Q82. Statement I: Evaluation uses testing data to assess model performance.
Statement II: Using training data for evaluation may cause overfitting.
a) Both are true
b) Both are false
c) I is true but II is false
d) I is false but II is true
Q83. Deployment is the:
a) First stage
b) Third stage
c) Final stage
d) Optional stage
Q84. The primary purpose of Deployment is to:
a) Make the AI solution ready to be used in the real world
b) Identify the stakeholders
c) Collect training data
d) Draw graphs
Q85. Which is a key step in the Deployment process?
a) Integration with existing systems
b) Selecting a theme
c) Identifying the 4Ws
d) Collecting newspaper evidence
Q86. Monitoring and maintenance are part of:
a) Deployment
b) Problem Scoping
c) Data Exploration
d) AI Ethics only
Q87. Under the “Human Rights” principle of AI Ethics, which of the following should be checked?
a) Whether AI increases company profits
b) Whether AI takes away freedom or discriminates against people
c) Whether AI is expensive to build
d) Whether AI uses cloud storage
Q88. “Bias” in AI Ethics often arises from:
a) The programming language used
b) The collected/training data not equally representing all sections of population
c) The internet speed
d) The number of developers involved
Q89. Under the “Privacy” principle of AI Ethics, a key concern is:
a) How fast the AI runs
b) Whether AI collects personal data and what it does with that data
c) The colour scheme of the AI app
d) The number of users of the AI
Q90. “Inclusion” as an AI Ethics principle means that AI:
a) Must benefit only rich people
b) Must not discriminate against any group, causing disadvantage
c) Should be difficult to use
d) Should only be available in cities
Q91. Statement I: Bias may originate from collected data.
Statement II: Bias in training data can appear in AI results.
a) Both are true
b) Both are false
c) I is true but II is false
d) I is false but II is true
Q92. Assertion (A): Data Acquisition is essential for an AI project that needs to make predictions.
Reason (R): AI systems need data for training.
a) Both A and R are true, and R is the correct explanation of A.
b) Both A and R are true, but R is not the correct explanation of A.
c) A is true, but R is false.
d) A is false, but R is true.
Q93. Assertion (A): Training data should be relevant to the problem statement.
Reason (R): Unrelated training data can lead to incorrect predictions.
a) Both A and R are true, and R is the correct explanation of A.
b) Both A and R are true, but R is not the correct explanation of A.
c) A is true, but R is false.
d) A is false, but R is true.
Q94. Assertion (A): Data Exploration is performed before Modelling.
Reason (R): Exploring data helps identify trends and patterns and can guide model selection.
a) Both A and R are true, and R is the correct explanation of A.
b) Both A and R are true, but R is not the correct explanation of A.
c) A is true, but R is false.
d) A is false, but R is true.
Q95. Assertion (A): Deep Learning is a subset of Machine Learning.
Reason (R): AI is the umbrella term, with ML under AI and DL under ML.
a) Both A and R are true, and R is the correct explanation of A.
b) Both A and R are true, but R is not the correct explanation of A.
c) A is true, but R is false.
d) A is false, but R is true.
