Artificial Intelligence (AI) MCQs

Solved Artificial Intelligence (AI) MCQs with answers and explanations. Free practice questions for PPSC, FPSC, CSS and other government job exams on FreeSekho.

1. 1. What are the three basic types of layers commonly found in a neural network?

   1. Input layer, hidden layer(s), and output layer ✓
   2. Only an input layer
   3. Only an output layer
   4. Storage layer, memory layer, and cache layer

2. 2. Which type of machine learning is commonly used for 'clustering' similar data points together?

   1. Unsupervised learning ✓
   2. Supervised learning
   3. Reinforcement learning
   4. Manual programming

3. 3. What type of data is a Recurrent Neural Network (RNN) particularly well-suited for processing?

   1. Sequential data, such as text or time series ✓
   2. Only static images
   3. Only audio hardware signals
   4. Only numerical constants

4. 4. What is a 'digital twin' in the context of AI and industry?

   1. A virtual, data-driven replica of a physical object, process, or system used for simulation and analysis ✓
   2. A physical backup copy of a computer
   3. A type of network cable
   4. A type of file compression format

5. 5. What is 'fine-tuning' in the context of machine learning models?

   1. The process of further training a pre-trained model on a specific, smaller dataset to specialize it for a particular task ✓
   2. The process of deleting a model entirely
   3. The process of physically repairing computer hardware
   4. The process of connecting to the internet

6. 6. What is 'labeling' data typically involve in supervised learning?

   1. Assigning correct, known answers or categories to data so a model can learn from them ✓
   2. Deleting unnecessary data from a dataset
   3. Compressing data to save storage space
   4. Encrypting data for security purposes

7. 7. Which IBM AI system famously won the quiz show 'Jeopardy!' against human champions in 2011?

   1. Watson ✓
   2. Deep Blue
   3. AlphaGo
   4. Siri

8. 8. What is 'multimodal AI'?

   1. AI systems capable of processing and understanding multiple types of input, such as text, images, and audio together ✓
   2. AI systems that can only process a single type of input
   3. A type of computer hardware component
   4. A type of network cable

9. 9. What does 'data privacy regulation' aim to protect in the context of AI and technology?

   1. An individual's personal information from being misused or accessed without proper consent ✓
   2. The physical hardware of a computer only
   3. The speed of a computer's processor
   4. The color scheme of a software interface

10. 10. What is 'robotics' in relation to AI?

    1. A field combining AI, mechanical engineering, and electronics to design and build robots capable of performing tasks ✓
    2. A field only concerned with computer hardware repair
    3. A field only concerned with internet connectivity
    4. A field only concerned with data storage

11. 11. What is the primary goal of 'optical character recognition' combined with AI in digitizing old documents?

    1. To convert scanned images of printed or handwritten text into searchable, editable digital text ✓
    2. To physically shred old paper documents
    3. To compress video files
    4. To connect a printer to Wi-Fi

12. 12. What is 'GPU acceleration' commonly used for in AI development?

    1. Using Graphics Processing Units to speed up the computationally intensive process of training AI models ✓
    2. Slowing down computer processing intentionally
    3. A type of network security protocol
    4. A type of file compression method

13. 13. What is the concept of 'human-in-the-loop' in AI systems?

    1. Designing AI systems so that human oversight and intervention remain part of the decision-making process ✓
    2. Removing all human involvement from AI systems entirely
    3. A type of computer hardware
    4. A type of network protocol

14. 14. What is 'pattern recognition' in the context of AI?

    1. The ability of a system to identify regularities, trends, or structures within data ✓
    2. The physical assembly of computer hardware components
    3. A type of network cable
    4. A type of file compression method

15. 15. Which mathematician and computer scientist is often referred to as the 'father of theoretical computer science and artificial intelligence'?

    1. Alan Turing ✓
    2. John McCarthy
    3. Bill Gates
    4. Charles Babbage

16. 16. What is meant by an AI model being 'pre-trained'?

    1. The model has already been trained on a large, general dataset before being fine-tuned or used for specific tasks ✓
    2. The model has never been trained on any data at all
    3. The model can only be used once and then must be discarded
    4. The model requires no data to function

17. 17. What is 'autonomous' generally understood to mean when describing a system, such as a vehicle or robot?

    1. Capable of operating and making decisions independently, with minimal or no human control ✓
    2. Requiring constant, direct human control for every single action
    3. A type of computer hardware component
    4. A type of network cable

18. 18. Most AI systems in use today, such as voice assistants and recommendation engines, are examples of which type of AI?

    1. Narrow AI ✓
    2. General AI
    3. Superintelligent AI
    4. Conscious AI

19. 19. What was one of the earliest formal fields of AI research, focused on mimicking the decision-making ability of a human expert in a specific domain?

    1. Expert systems ✓
    2. Neural networks
    3. Robotics
    4. Computer vision

20. 20. Which early period (roughly the 1950s-1960s) is often considered the birth of AI as a formal academic field?

    1. The 1950s, following the Dartmouth Conference in 1956 ✓
    2. The 1990s
    3. The 2010s
    4. The 1800s

21. 21. What is the purpose of a 'CAPTCHA' test commonly seen on websites?

    1. To distinguish between human users and automated bots/programs ✓
    2. To speed up a website's loading time
    3. To compress images on a webpage
    4. To translate a webpage into another language

22. 22. What is 'Natural Language Processing' (NLP)?

    1. A branch of AI focused on enabling computers to understand, interpret, and generate human language ✓
    2. A branch of AI focused only on image recognition
    3. A type of computer hardware
    4. A type of network protocol

23. 23. What is the general relationship between the amount and quality of training data and the performance of most machine learning models?

    1. Generally, more relevant, high-quality data tends to improve a model's performance, up to a point ✓
    2. The amount and quality of data has no effect on model performance whatsoever
    3. Less data always produces better results than more data
    4. Only the quantity of data matters, and quality is irrelevant

24. 24. What is a 'chat interface' in the context of AI assistants like chatbots?

    1. A text-based (or voice-based) interface that allows users to interact conversationally with an AI system ✓
    2. A type of computer hardware component
    3. A type of network cable
    4. A type of file storage format

25. 25. What is a 'ChatGPT-style' AI assistant primarily built on, at a technical level?

    1. A large language model trained on vast amounts of text data ✓
    2. A simple rule-based lookup table with no learning involved
    3. A type of computer hardware component
    4. A type of network cable

26. 26. Why is it generally recommended that users verify information generated by AI chatbots?

    1. Because AI models can sometimes produce inaccurate or fabricated information ('hallucinations') ✓
    2. Because AI models are always 100% accurate with no exceptions
    3. Because verification is legally required for every single query
    4. Because AI chatbots cannot process any factual information at all

27. 27. What is a 'Large Language Model' (LLM)?

    1. A type of AI model trained on vast amounts of text data to understand and generate human-like language ✓
    2. A type of computer hardware
    3. A type of database only used for numbers
    4. A type of network cable

28. 28. What does 'GAN' stand for in the context of AI?

    1. Generative Adversarial Network ✓
    2. General Artificial Network
    3. Global Access Network
    4. Graphic Analysis Network

29. 29. What is an 'API' (as commonly used to access AI services)?

    1. Application Programming Interface — a set of protocols that allows different software applications to communicate with each other ✓
    2. A type of computer hardware
    3. A type of network cable
    4. A type of file extension

30. 30. Which AI technique is commonly used by self-driving cars to detect and interpret their surroundings?

    1. Computer Vision ✓
    2. Word processing
    3. Spreadsheet calculation
    4. Basic text editing

31. 31. What is 'clustering' in unsupervised machine learning?

    1. Grouping similar data points together based on shared characteristics, without predefined labels ✓
    2. Randomly separating data points with no pattern
    3. A type of computer hardware
    4. A type of network protocol

32. 32. What is 'fraud detection' in the context of AI applications in finance?

    1. Using AI algorithms to identify unusual or suspicious transaction patterns that may indicate fraudulent activity ✓
    2. Manually checking every single transaction by hand only
    3. A type of computer hardware
    4. A type of network cable

33. 33. What is the significance of the phrase 'garbage in, garbage out' as it relates to AI and machine learning?

    1. It highlights that poor-quality or flawed input data will lead to poor-quality or flawed model outputs ✓
    2. It refers to a method of physically disposing of old computer hardware
    3. It refers to a type of network cable installation
    4. It has no relevance to AI or machine learning

34. 34. What is 'personalization' in the context of AI-driven digital experiences?

    1. Tailoring content, recommendations, or services to an individual user's preferences and behavior ✓
    2. Providing the exact same experience to every single user with no variation
    3. A type of computer hardware component
    4. A type of network protocol

35. 35. What is the fundamental difference between traditional rule-based programming and machine learning?

    1. In traditional programming, rules are explicitly coded by humans; in machine learning, the system learns rules/patterns from data ✓
    2. There is no difference between the two approaches
    3. Machine learning requires no data of any kind
    4. Traditional programming always requires more computing power than machine learning

36. 36. What is the connection between the name 'CAPTCHA' and the 'Turing Test'?

    1. CAPTCHA is essentially a reverse Turing Test, where a computer tries to determine if the user is human ✓
    2. There is no connection between the two concepts at all
    3. CAPTCHA was invented before the Turing Test existed
    4. CAPTCHA and the Turing Test are the exact same test

37. 37. What was the name of the first industrial robot, introduced in the early 1960s and used on a General Motors assembly line?

    1. Unimate ✓
    2. ASIMO
    3. Sophia
    4. Roomba

38. 38. What does the term 'AI assistant' most commonly refer to in everyday consumer technology?

    1. A software program that uses AI to help users perform tasks, answer questions, or manage schedules ✓
    2. A physical human assistant hired through an agency
    3. A type of computer hardware component only
    4. A type of network cable

39. 39. What is the purpose of the 'hidden layer(s)' in a neural network?

    1. To process and transform the input data through weighted connections before producing an output ✓
    2. To store the final results permanently
    3. To connect the computer to the internet
    4. To display the results visually

40. 40. What is 'Computer Vision' in AI?

    1. A field of AI that enables computers to interpret and understand visual information from the world, such as images and videos ✓
    2. A type of computer monitor
    3. A type of network protocol
    4. A type of file compression

41. 41. Which Japanese company developed the humanoid robot named ASIMO?

    1. Honda ✓
    2. Toyota
    3. Sony
    4. Panasonic

42. 42. Who proposed the famous 'Turing Test' to evaluate a machine's ability to exhibit intelligent behavior?

    1. Alan Turing ✓
    2. John McCarthy
    3. Marvin Minsky
    4. Geoffrey Hinton

43. 43. Which company developed the AlphaGo program that defeated a world champion Go player?

    1. DeepMind (a subsidiary of Google/Alphabet) ✓
    2. IBM
    3. Microsoft
    4. Apple

44. 44. In which year was the term 'Artificial Intelligence' coined at a famous academic conference?

    1. 1956 ✓
    2. 1936
    3. 1975
    4. 1990

45. 45. What is a common ethical concern regarding AI and employment?

    1. The potential for AI and automation to displace certain human jobs ✓
    2. AI has no impact on employment whatsoever
    3. AI can only create jobs and never displace any
    4. AI is banned from all workplaces globally

46. 46. What is a Convolutional Neural Network (CNN) primarily used for?

    1. Processing and analyzing visual data, such as images ✓
    2. Processing only numerical spreadsheets
    3. Managing email systems
    4. Connecting to the internet

47. 47. What is 'bias' in the context of AI systems?

    1. Systematic errors or unfair outcomes in an AI system, often resulting from biased training data ✓
    2. A type of computer hardware component
    3. A type of network protocol
    4. A type of file extension

48. 48. What is 'predictive maintenance', an AI application used in manufacturing and industry?

    1. Using AI to predict when equipment is likely to fail so maintenance can be performed proactively ✓
    2. Waiting for equipment to completely break down before any action is taken
    3. A method of manually inspecting equipment daily with no data analysis
    4. A type of computer hardware component

49. 49. What is the role of 'parameters' in a large machine learning model?

    1. Internal variables the model learns and adjusts during training to make accurate predictions ✓
    2. The physical size of the computer running the model
    3. The brand name of the hardware used
    4. The color scheme of the software interface

50. 50. What is a 'smart home' device, and how does it typically relate to AI?

    1. A home device (like a smart thermostat or speaker) that uses AI to learn user preferences and automate tasks ✓
    2. A home built only with traditional bricks and no technology
    3. A type of computer hardware component only
    4. A type of network cable only

51. 51. What is an 'activation function' in a neural network used for?

    1. To determine whether a neuron should be activated based on its input, introducing non-linearity ✓
    2. To physically power the computer
    3. To store data permanently
    4. To connect to the internet

52. 52. Why is data privacy an important concern in the development of AI systems?

    1. Because AI systems often process large amounts of personal data, raising risks of misuse or unauthorized access ✓
    2. Because AI systems never use any personal data
    3. Because data privacy is unrelated to AI
    4. Because AI systems only use publicly available government data

53. 53. What does 'inference' mean in the context of a trained AI model?

    1. The process of using a trained model to make predictions on new, unseen data ✓
    2. The process of initially training a model from scratch
    3. A type of computer hardware
    4. A type of network protocol

54. 54. What is a 'dataset' in the context of AI and machine learning?

    1. A structured collection of data used for training, testing, or evaluating a model ✓
    2. A single isolated data point
    3. A type of computer hardware
    4. A type of network protocol

55. 55. What is 'zero-shot learning'?

    1. A model's ability to perform a task it has not explicitly been trained on, using general knowledge alone ✓
    2. A model that requires extensive retraining for every new task
    3. A type of computer hardware
    4. A type of network cable

56. 56. Who is generally credited with coining the term 'Artificial Intelligence'?

    1. John McCarthy ✓
    2. Alan Turing
    3. Bill Gates
    4. Steve Jobs

57. 57. What is 'anomaly detection' in AI, and where is it commonly applied?

    1. Identifying unusual patterns that do not conform to expected behavior, commonly used in fraud detection and cybersecurity ✓
    2. Detecting only normal, expected patterns in data
    3. A type of computer hardware component
    4. A type of network cable

58. 58. What does 'CAPTCHA' stand for?

    1. Completely Automated Public Turing test to tell Computers and Humans Apart ✓
    2. Computer Automated Program to Test Computer Hardware Access
    3. Central Access Protocol To Control Hardware Applications
    4. Coded Automated Password Test for Computer Hardware Access

59. 59. What is a Generative Adversarial Network (GAN) generally used for?

    1. Generating new, realistic data (such as images) by having two neural networks compete against each other ✓
    2. Only storing data
    3. Only connecting to the internet
    4. Only printing documents

60. 60. What is 'sentiment analysis' commonly used for by businesses?

    1. To gauge public or customer opinion from text data such as reviews or social media posts ✓
    2. To physically manufacture products
    3. To manage employee payroll
    4. To design office furniture

61. 61. What is a 'confusion matrix' used for in evaluating a classification model?

    1. To visualize the performance of a model by showing correct and incorrect predictions across categories ✓
    2. To physically confuse a computer's hardware
    3. A type of network security tool
    4. A type of file compression tool

62. 62. What is a 'feature' in the context of a machine learning dataset?

    1. An individual measurable property or characteristic of the data used as input for a model ✓
    2. The final output or prediction of a model
    3. A type of computer hardware
    4. A type of network cable

63. 63. What is the term for an AI system continuing to learn and adapt after its initial deployment, based on new data?

    1. Continual (or online) learning ✓
    2. Static learning
    3. One-time learning
    4. Frozen learning

64. 64. Why are deepfakes considered a potential societal risk?

    1. Because they can be used to spread misinformation or impersonate individuals convincingly ✓
    2. Because they always improve video quality with no downsides
    3. Because they are illegal to create anywhere in the world
    4. Because they only affect computer hardware performance

65. 65. What is an AI-powered 'image generator' capable of doing?

    1. Creating new images based on text descriptions or other input prompts ✓
    2. Only displaying pre-existing images without any changes
    3. Only compressing image file sizes
    4. Only printing physical photographs

66. 66. What term is used to describe periods of reduced funding and interest in AI research, historically occurring after periods of unmet expectations?

    1. AI Winter ✓
    2. AI Freeze
    3. AI Recession
    4. AI Drought

67. 67. What is the significance of 'context window' size in a language model?

    1. It determines how much text (input and conversation history) the model can consider at once when generating a response ✓
    2. It refers to the physical size of the computer screen
    3. It refers to the number of internet browser tabs open
    4. It refers to the color settings of the display

68. 68. What is 'General AI' (also called 'Strong AI' or AGI)?

    1. A hypothetical AI with the ability to understand, learn, and apply intelligence across a wide range of tasks like a human
    2. AI limited to a single specific task ✓
    3. A type of search engine
    4. A type of antivirus software

69. 69. What is 'transfer learning' in machine learning?

    1. A technique where a model trained on one task is reused or adapted as the starting point for a related task ✓
    2. Physically transferring a computer from one location to another
    3. Deleting a trained model entirely and starting over
    4. A type of network data transfer protocol

70. 70. What does 'accuracy' measure in the context of evaluating a machine learning model?

    1. The proportion of correct predictions made by the model out of all predictions ✓
    2. The physical speed of the computer running the model
    3. The amount of storage space a model requires
    4. The internet connection speed used during training

71. 71. How is AI applied in the field of translation services (such as Google Translate)?

    1. Using machine learning models to automatically translate text or speech between languages ✓
    2. By manually looking up each word in a printed dictionary
    3. By physically connecting two countries via cable
    4. By using only basic word-for-word substitution with no learning involved

72. 72. What is 'crowdsourcing' data labeling, sometimes used to prepare training data for AI?

    1. Gathering labeled data contributions from a large number of distributed people, often via the internet ✓
    2. Generating labeled data using only a single expert
    3. A type of computer hardware component
    4. A type of network cable

73. 73. What does 'speech-to-text' technology do, as opposed to text-to-speech?

    1. Converts spoken audio into written text ✓
    2. Converts written text into spoken audio
    3. Compresses text files
    4. Connects a computer to Wi-Fi

74. 74. What is 'algorithmic bias'?

    1. Unfair or prejudiced outcomes produced by an AI system due to biased training data or flawed design ✓
    2. A hardware defect in a computer processor
    3. A type of network security protocol
    4. A type of file compression error

75. 75. What does 'AI ethics' broadly concern itself with?

    1. The moral principles and responsible practices guiding the development and use of AI systems ✓
    2. Only the physical design of computer hardware
    3. Only the speed of internet connections
    4. Only the price of AI software

76. 76. What is an 'AI agent'?

    1. A system that perceives its environment and takes actions to achieve specific goals ✓
    2. A physical robot only
    3. A type of computer virus
    4. A type of storage device

77. 77. What is 'predictive text' (as seen on smartphone keyboards) an example of?

    1. An AI/machine learning feature that suggests the next word or completes words as a user types ✓
    2. A hardware component of the keyboard
    3. A type of network protocol
    4. A type of antivirus software

78. 78. What is 'Artificial Superintelligence' (ASI)?

    1. A hypothetical form of AI that surpasses human intelligence across virtually all fields ✓
    2. AI that performs only basic calculations
    3. A type of computer virus
    4. A type of hardware chip

79. 79. Which of these best distinguishes 'automation' from 'AI'?

    1. Automation follows fixed, pre-set rules, while AI can learn, adapt, and make decisions based on data ✓
    2. Automation and AI are exactly the same concept with no distinction
    3. AI always requires more manual programming than automation
    4. Automation always requires machine learning while AI never does

80. 80. How is AI commonly used in the banking and finance industry?

    1. For fraud detection, credit risk assessment, and algorithmic trading ✓
    2. Only for printing physical currency
    3. Only for building bank branches
    4. Only for manual paper record keeping

81. 81. What is one key advantage of using synthetic data in AI training?

    1. It can help protect privacy and supplement limited real-world data ✓
    2. It always produces worse results than real data with no exceptions
    3. It cannot be used for any type of model training
    4. It requires no computing resources to generate

82. 82. What is 'natural language understanding' (NLU), a subfield of NLP?

    1. The ability of an AI system to comprehend and interpret the meaning of human language input ✓
    2. The process of generating random text with no meaning
    3. A type of computer hardware
    4. A type of network protocol

83. 83. What is 'cloud computing' as it commonly relates to AI development?

    1. Using remote servers hosted on the internet to store, manage, and process data, often used to train and run AI models ✓
    2. Storing data only on a local computer with no internet involved
    3. A type of computer hardware component
    4. A type of network cable

84. 84. What does 'CNN' commonly stand for in the context of deep learning?

    1. Convolutional Neural Network ✓
    2. Central Node Network
    3. Computer Neural Network
    4. Combined Network Node

85. 85. What is a 'spam filter' in email, and what technology commonly powers it?

    1. A tool that automatically identifies and separates unwanted emails, often powered by machine learning ✓
    2. A physical filter installed in computer hardware
    3. A tool that increases internet speed
    4. A tool that compresses email attachments

86. 86. How is Machine Learning related to Artificial Intelligence?

    1. Machine Learning is a subset (branch) of Artificial Intelligence ✓
    2. Machine Learning is a completely separate field from AI
    3. Artificial Intelligence is a subset of Machine Learning
    4. They are unrelated fields

87. 87. What is the primary purpose of combining OCR with AI in modern applications?

    1. To accurately extract and interpret text from scanned images or documents ✓
    2. To compress video files
    3. To increase internet bandwidth
    4. To cool computer hardware

88. 88. What does 'AI' stand for?

    1. Artificial Intelligence ✓
    2. Automated Interaction
    3. Artificial Interface
    4. Automated Intelligence

89. 89. Who is often called the 'father of Artificial Intelligence' for coining the term and organizing the Dartmouth Conference?

    1. John McCarthy ✓
    2. Alan Turing
    3. Steve Jobs
    4. Tim Berners-Lee

90. 90. What does 'Big Data' refer to?

    1. Extremely large and complex datasets that traditional data-processing methods struggle to handle ✓
    2. A small dataset used for testing
    3. A type of computer virus
    4. A type of hardware chip

91. 91. How is AI commonly used in the field of education?

    1. For personalized learning platforms, automated grading, and virtual tutoring ✓
    2. Only for building school infrastructure
    3. Only for printing textbooks
    4. Only for manual attendance taking

92. 92. What is 'unsupervised learning' in machine learning?

    1. A type of learning where the model finds patterns or structures in data without labeled outputs ✓
    2. A type of learning that always requires labeled data
    3. A type of learning used only in robotics
    4. A type of learning with no data involved

93. 93. How is Deep Learning related to Machine Learning?

    1. Deep Learning is a subset (specialized branch) of Machine Learning ✓
    2. Deep Learning and Machine Learning are unrelated
    3. Machine Learning is a subset of Deep Learning
    4. They are the same thing with no distinction

94. 94. What is a common criticism or limitation of current AI systems regarding true understanding?

    1. AI systems often recognize patterns in data without genuinely 'understanding' meaning the way humans do ✓
    2. AI systems always have complete human-like consciousness
    3. AI systems require no data to operate
    4. AI systems can never process any form of language

95. 95. How is AI used in weather forecasting?

    1. To analyze large amounts of atmospheric data and improve the accuracy of weather predictions ✓
    2. Only to physically measure rainfall with buckets
    3. Only to build weather stations
    4. Only to print weather reports

96. 96. What is 'image classification' in computer vision?

    1. The task of assigning a label or category to an entire image based on its content ✓
    2. The task of physically printing an image
    3. A type of computer hardware
    4. A type of network protocol

97. 97. What does 'AI-generated content' generally refer to?

    1. Text, images, audio, or video created by an artificial intelligence system rather than directly by a human ✓
    2. Content that is always factually perfect with no errors
    3. Content that can only be created by human writers
    4. Content that requires no computing resources to produce

98. 98. What is the basic processing unit of an artificial neural network called?

    1. A neuron (or node) ✓
    2. A byte
    3. A pixel
    4. A register

99. 99. What is 'facial recognition' technology?

    1. A technology that identifies or verifies a person's identity using their facial features ✓
    2. A technology used only to compress images
    3. A type of computer virus
    4. A type of hardware port

100. 100. What is the name of the humanoid robot developed by Hanson Robotics that gained notable public attention and was granted honorary citizenship by Saudi Arabia in 2017?

     1. Sophia ✓
     2. ASIMO
     3. Pepper
     4. Atlas

101. 101. What is 'responsible AI' generally understood to mean?

     1. Developing and deploying AI systems in a way that is ethical, fair, transparent, and accountable ✓
     2. Developing AI systems as quickly as possible with no other considerations
     3. A type of computer hardware certification
     4. A type of network security protocol

102. 102. What is a 'self-driving car' (autonomous vehicle) an example of?

     1. The application of AI, including computer vision and sensors, to navigate and drive without human control ✓
     2. A car that runs only on manual controls
     3. A type of computer hardware component
     4. A type of network protocol

103. 103. What is 'Machine Learning' (ML)?

     1. A subset of AI where systems learn patterns from data to improve performance without being explicitly programmed for every rule ✓
     2. A type of computer hardware
     3. A type of network protocol
     4. A method of manually programming every possible outcome

104. 104. What is a 'virtual assistant' (such as Siri, Alexa, or Google Assistant)?

     1. An AI-powered software agent that can perform tasks or services for a user based on commands or questions ✓
     2. A type of physical hardware device only
     3. A type of computer virus
     4. A type of network cable

105. 105. What is 'semi-supervised learning'?

     1. A learning approach that uses a combination of a small amount of labeled data and a large amount of unlabeled data ✓
     2. A learning approach that uses only labeled data
     3. A learning approach that uses only unlabeled data
     4. A learning approach with no data at all

106. 106. What is one major benefit of using AI-powered translation tools for everyday users?

     1. They can quickly translate text or speech between many languages, aiding communication ✓
     2. They can physically print translated documents automatically
     3. They eliminate the need for the internet entirely
     4. They can only translate a single fixed language pair with no flexibility

107. 107. How is AI used in social media platforms to enhance user experience?

     1. To curate personalized content feeds and suggest connections based on user behavior ✓
     2. Only to physically manufacture smartphones
     3. Only to provide electricity to servers
     4. Only to design company logos

108. 108. What is 'sentiment analysis', a common NLP application?

     1. Using AI to determine the emotional tone (positive, negative, or neutral) behind a piece of text ✓
     2. A method of physically measuring emotions
     3. A type of computer hardware
     4. A type of network protocol

109. 109. What is 'automation' in the broader context of AI and technology?

     1. The use of technology to perform tasks with minimal human intervention ✓
     2. The manual completion of all tasks by humans
     3. A type of computer hardware
     4. A type of file format

110. 110. What is a 'recommendation algorithm' generally based on?

     1. Analyzing patterns in user behavior and preferences to suggest relevant content or products ✓
     2. Randomly selecting content with no analysis involved
     3. A type of computer hardware
     4. A type of network cable

111. 111. What is 'Generative AI'?

     1. AI systems capable of creating new content, such as text, images, or audio, based on learned patterns ✓
     2. AI that can only analyze existing data without creating anything new
     3. A type of computer hardware
     4. A type of network protocol

112. 112. What is the purpose of 'model evaluation' in the machine learning development process?

     1. To assess how well a trained model performs on new, unseen data before deployment ✓
     2. To physically assemble the computer hardware running the model
     3. To connect the model to a printer
     4. To compress the model's file size only

113. 113. In what year did Alan Turing propose his famous test, in a paper titled 'Computing Machinery and Intelligence'?

     1. 1950 ✓
     2. 1940
     3. 1960
     4. 1936

114. 114. How is AI commonly used in the healthcare industry?

     1. For tasks such as analyzing medical images, assisting in diagnosis, and predicting disease risk ✓
     2. Only for scheduling staff shifts manually
     3. Only for printing prescriptions
     4. Only for physical building maintenance

115. 115. What is a 'weight' in the context of a neural network?

     1. A parameter that determines the strength or importance of a connection between neurons ✓
     2. A physical measurement of the computer's hardware
     3. A type of file format
     4. A type of network cable

116. 116. How is AI commonly used in agriculture?

     1. For tasks such as monitoring crop health, predicting yields, and enabling precision farming ✓
     2. Only for manually plowing fields
     3. Only for building barns
     4. Only for transporting livestock

117. 117. How might AI be used in customer service by many companies today?

     1. Through AI-powered chatbots that can answer common customer questions automatically ✓
     2. Only through in-person visits to company offices
     3. Only through printed customer manuals
     4. Only through postal mail correspondence

118. 118. Why might diverse and representative training data be important for reducing bias in AI systems?

     1. Because models trained on non-diverse data may perform poorly or unfairly for underrepresented groups ✓
     2. Diverse data has no effect on AI model performance or fairness
     3. AI models never require any training data
     4. Training data diversity only affects hardware performance

119. 119. What is 'edge AI' or 'edge computing' in relation to AI?

     1. Running AI processing directly on local devices (like smartphones) rather than relying solely on remote cloud servers ✓
     2. Running AI only on large, centralized supercomputers
     3. A type of computer hardware component with no processing ability
     4. A type of network cable

120. 120. What is 'reinforcement learning'?

     1. A type of learning where an agent learns to make decisions by receiving rewards or penalties for its actions ✓
     2. A type of learning that only uses labeled data
     3. A type of learning that requires no interaction with an environment
     4. A type of learning used only for image recognition

121. 121. What is a 'chatbot'?

     1. A computer program designed to simulate human conversation, often using AI ✓
     2. A type of physical robot only
     3. A type of computer virus
     4. A type of hardware component

122. 122. Which company developed the Google Assistant virtual assistant?

     1. Google ✓
     2. Apple
     3. Amazon
     4. Microsoft

123. 123. Why might it be important to disclose when content has been generated or significantly assisted by AI?

     1. To maintain transparency and allow audiences to evaluate the content appropriately ✓
     2. Because AI-generated content is always illegal
     3. Because disclosure is never necessary in any context
     4. Because AI-generated content cannot be viewed by humans

124. 124. What is a 'humanoid robot'?

     1. A robot designed to resemble and mimic human body structure and movement ✓
     2. A robot that looks like a car
     3. A type of computer hardware component
     4. A type of network protocol

125. 125. What is a neural network's 'input layer' responsible for?

     1. Receiving the initial raw data that will be processed by the network ✓
     2. Producing the final output of the network
     3. Storing the network's data permanently
     4. Connecting the network to the internet

126. 126. What is a 'knowledge cutoff' in the context of AI language models?

     1. The point in time after which a model has no information about events, since its training data has a fixed endpoint ✓
     2. A hardware limitation on how much data a computer can store
     3. A type of network security feature
     4. A type of file compression limit

127. 127. What is the purpose of the Turing Test?

     1. To determine whether a machine can exhibit behavior indistinguishable from a human ✓
     2. To measure a computer's processing speed
     3. To test internet connection speed
     4. To measure a computer's storage capacity

128. 128. What does 'OCR', often used alongside AI, stand for?

     1. Optical Character Recognition ✓
     2. Online Content Recognition
     3. Optical Computer Reading
     4. Original Character Recording

129. 129. What is 'tokenization' in the context of natural language processing?

     1. The process of breaking down text into smaller units, such as words or subwords, for a model to process ✓
     2. The process of encrypting a file for security
     3. A type of computer hardware
     4. A type of network protocol

130. 130. What is Artificial Intelligence broadly defined as?

     1. The simulation of human intelligence processes by machines, especially computer systems ✓
     2. A type of computer hardware
     3. A type of network cable
     4. A file storage format

131. 131. What does 'AI transparency' generally refer to?

     1. The degree to which the decision-making process of an AI system can be understood and explained ✓
     2. The physical transparency of computer hardware casings
     3. The speed of an AI system's internet connection
     4. The cost of purchasing AI software

132. 132. What is a common concern regarding the environmental impact of training large AI models?

     1. The significant amount of energy and computational resources required, which can have a notable carbon footprint ✓
     2. AI models require no energy to train at all
     3. Training AI models has zero environmental impact
     4. AI models are always trained using only renewable energy globally

133. 133. Who developed the ELIZA chatbot program in the 1960s?

     1. Joseph Weizenbaum ✓
     2. Alan Turing
     3. John McCarthy
     4. Bill Gates

134. 134. What does the term 'AI hallucination' refer to in the context of generative AI models?

     1. When an AI model generates false or nonsensical information that is presented as if it were factual ✓
     2. When an AI model runs out of battery power
     3. A type of computer hardware malfunction
     4. A type of network connectivity issue

135. 135. What is a common real-world example of a clustering application?

     1. Grouping customers with similar purchasing behavior for targeted marketing ✓
     2. Printing a physical document
     3. Connecting to a Wi-Fi network
     4. Formatting a hard disk

136. 136. What is 'training data' in the context of machine learning?

     1. The dataset used to teach a machine learning model to make predictions or decisions ✓
     2. The final output produced by a model
     3. A type of computer hardware
     4. A type of network cable

137. 137. What is a 'decision tree' in machine learning?

     1. A model that makes decisions by splitting data into branches based on feature values, resembling a tree structure ✓
     2. A physical tree used to store computer hardware
     3. A type of network topology only
     4. A type of file format

138. 138. What is 'backpropagation' in the context of training neural networks?

     1. An algorithm used to adjust the weights of a neural network by propagating the error backward through the network ✓
     2. A method of physically rewiring computer hardware
     3. A type of internet connection
     4. A type of file compression

139. 139. What is a common example of a 'multimodal' AI capability?

     1. An AI system that can analyze both an image and a text description together to answer a question ✓
     2. An AI system that can only read plain text with no other input types
     3. A type of computer hardware upgrade
     4. A type of network protocol

140. 140. What does 'API' stand for?

     1. Application Programming Interface ✓
     2. Automated Program Instruction
     3. Advanced Processing Interface
     4. Applied Program Integration

141. 141. What is the 'Turing Award' often considered equivalent to in the field of computing?

     1. The 'Nobel Prize' of computer science ✓
     2. An Olympic medal
     3. A business award only for startups
     4. A hardware design patent

142. 142. What is 'synthetic data' in the context of machine learning?

     1. Artificially generated data used to train models, often when real data is scarce or sensitive ✓
     2. Data collected directly from real-world human interactions only
     3. A type of computer hardware
     4. A type of network protocol

143. 143. What is 'speech recognition' technology used for?

     1. Converting spoken language into text or commands that a computer can process ✓
     2. Converting text into printed documents
     3. Compressing audio files
     4. Connecting to the internet

144. 144. Which of the following is an example of Generative AI capability?

     1. Creating a new image or piece of text from a text prompt ✓
     2. Only sorting numbers in a spreadsheet
     3. Only connecting to Wi-Fi
     4. Only printing documents

145. 145. What is the term for AI systems designed to detect and block malicious software or cyberattacks using learned patterns?

     1. AI-driven threat detection ✓
     2. Manual paper-based security logs
     3. A type of computer hardware component
     4. A type of network cable

146. 146. Which three researchers are often referred to as 'Godfathers of AI' and jointly received the 2018 Turing Award for their work on deep learning?

     1. Geoffrey Hinton, Yann LeCun, and Yoshua Bengio ✓
     2. Alan Turing, John McCarthy, and Marvin Minsky
     3. Bill Gates, Steve Jobs, and Elon Musk
     4. Larry Page, Sergey Brin, and Mark Zuckerberg

147. 147. What is 'object detection' in computer vision?

     1. The task of identifying and locating specific objects within an image or video ✓
     2. The task of deleting objects from an image
     3. A type of computer hardware component
     4. A type of network cable

148. 148. Which company developed the Alexa virtual assistant?

     1. Amazon ✓
     2. Apple
     3. Google
     4. Microsoft

149. 149. What is a common application of AI in cybersecurity?

     1. Detecting unusual network activity or potential threats by analyzing patterns in data ✓
     2. Only manufacturing computer hardware
     3. Only designing company websites
     4. Only printing security badges

150. 150. What is a 'drone' commonly equipped with AI capable of doing?

     1. Autonomously navigating, capturing imagery, or performing tasks without constant manual human control ✓
     2. Only functioning as a stationary display object
     3. Only functioning as a type of computer hardware storage
     4. Only functioning as a network cable

151. 151. What is a 'model' in the context of machine learning?

     1. A mathematical representation trained on data to make predictions or decisions ✓
     2. A physical piece of computer hardware
     3. A type of internet browser
     4. A type of operating system

152. 152. Which IBM computer famously defeated world chess champion Garry Kasparov in 1997?

     1. Deep Blue ✓
     2. Watson
     3. AlphaGo
     4. ENIAC

153. 153. What is an 'algorithm' in the context of AI and computing?

     1. A step-by-step set of rules or instructions followed to solve a problem or complete a task ✓
     2. A type of computer hardware
     3. A type of internet connection
     4. A type of file format

154. 154. What is an 'expert system' in AI?

     1. A computer system designed to emulate the decision-making ability of a human expert in a specific field ✓
     2. A type of hardware processor
     3. A type of network cable
     4. A type of file format

155. 155. In reinforcement learning, what is a 'reward'?

     1. A positive feedback signal given to an agent for taking a beneficial action ✓
     2. A type of computer hardware
     3. A type of labeled dataset
     4. A type of neural network layer

156. 156. What is 'underfitting' in machine learning?

     1. When a model is too simple to capture the underlying patterns in the data, leading to poor performance ✓
     2. When a model performs perfectly on all data
     3. When a model has learned the training data too well
     4. When a model has too much data

157. 157. What is a 'recommendation system' (as used by platforms like Netflix or Amazon)?

     1. An AI system that suggests content or products to users based on their preferences and behavior ✓
     2. A system that only displays random content
     3. A type of computer hardware
     4. A type of network cable

158. 158. Which company developed the Siri virtual assistant integrated into its devices?

     1. Apple ✓
     2. Google
     3. Amazon
     4. Microsoft

159. 159. What is a 'benchmark' in the context of evaluating AI models?

     1. A standardized test or dataset used to measure and compare the performance of different models ✓
     2. A type of computer hardware component
     3. A type of network cable
     4. A type of file compression format

160. 160. What is a 'prompt' in the context of using generative AI tools like chatbots?

     1. The input text or instruction given by a user to guide the AI's response ✓
     2. A type of computer hardware
     3. A type of network cable
     4. A type of antivirus software

161. 161. In general terms, what tends to happen to a language model's capability as the number of its parameters increases (given sufficient quality data)?

     1. It often becomes capable of more complex and nuanced tasks, though with diminishing returns and higher cost ✓
     2. It always becomes less capable and less accurate
     3. The number of parameters has no relationship to capability
     4. It always becomes cheaper and faster to run with more parameters

162. 162. What is 'Deep Learning'?

     1. A subset of Machine Learning that uses multi-layered artificial neural networks to learn from large amounts of data ✓
     2. A type of database management system
     3. A method of manually coding every rule
     4. A type of computer hardware

163. 163. What does 'text-to-speech' (TTS) technology do?

     1. Converts written text into spoken audio output ✓
     2. Converts spoken audio into written text
     3. Compresses audio files
     4. Connects a computer to the internet

164. 164. What does 'model training' in machine learning generally involve?

     1. Feeding data into an algorithm so it can learn patterns and adjust its internal parameters ✓
     2. Physically assembling computer hardware components
     3. Connecting a computer to a printer
     4. Formatting a hard disk drive

165. 165. What is 'predictive analytics'?

     1. The use of data, statistical algorithms, and machine learning to identify the likelihood of future outcomes ✓
     2. The analysis of only past events with no forecasting involved
     3. A type of computer hardware
     4. A type of network protocol

166. 166. Which type of machine learning is commonly associated with training AI to play games, such as through trial and error?

     1. Reinforcement learning ✓
     2. Supervised learning
     3. Unsupervised learning
     4. Rule-based programming

167. 167. How do streaming platforms like Netflix typically use AI?

     1. To recommend movies and shows based on a user's viewing history and preferences ✓
     2. To physically manufacture DVDs
     3. To provide internet connectivity
     4. To repair computer hardware

168. 168. What is a voice assistant's 'wake word' (such as 'Hey Siri' or 'Alexa') used for?

     1. To activate the assistant and signal that a voice command will follow ✓
     2. To permanently shut down the device
     3. To disconnect the device from the internet
     4. To delete the device's stored data

169. 169. What is the general purpose of a 'validation set' in machine learning model development?

     1. To tune model parameters and check performance during training, separate from the final test set ✓
     2. To permanently store the final trained model
     3. A type of computer hardware
     4. A type of network protocol

170. 170. What is a 'deepfake'?

     1. A synthetic media file, often video or audio, generated by AI to convincingly depict someone saying or doing something they did not actually do ✓
     2. A genuine, unedited video recording
     3. A type of computer hardware
     4. A type of network cable

171. 171. What does the term 'AI-powered' generally imply about a product or service?

     1. That the product or service incorporates artificial intelligence techniques to enhance its functionality ✓
     2. That the product runs without any electricity
     3. That the product is entirely built by hand with no computers involved
     4. That the product has no software component at all

172. 172. What does 'AI literacy' refer to for the general public?

     1. A basic understanding of what AI is, how it works, and its capabilities and limitations ✓
     2. The ability to write complex AI programming code only
     3. A type of computer hardware certification
     4. A type of network security clearance

173. 173. What role does AI commonly play in email services?

     1. Filtering spam and organizing emails into categories ✓
     2. Physically delivering printed mail
     3. Manufacturing computer hardware
     4. Providing electricity to the building

174. 174. What is 'data mining'?

     1. The process of discovering patterns and useful information from large datasets ✓
     2. The physical extraction of metals from the earth
     3. A type of computer virus
     4. A type of network cable

175. 175. What is 'supervised learning' in machine learning?

     1. A type of learning where the model is trained on labeled data with known correct outputs ✓
     2. A type of learning where the model receives no data at all
     3. A type of learning that requires no training
     4. A type of learning used only for robots

176. 176. What is the 'output layer' of a neural network responsible for?

     1. Producing the final result or prediction of the network ✓
     2. Receiving the initial raw input data
     3. Physically cooling the computer hardware
     4. Connecting to an external network

177. 177. What is a key difference between a simple rule-based chatbot and an AI-powered conversational assistant using an LLM?

     1. An LLM-based assistant can generate more flexible, context-aware responses rather than following only fixed, pre-written scripts ✓
     2. There is no difference between the two types of systems
     3. A rule-based chatbot always understands language better than an LLM
     4. LLM-based assistants require no data or training of any kind

178. 178. What does 'AGI' stand for?

     1. Artificial General Intelligence ✓
     2. Automated General Interface
     3. Artificial Global Intelligence
     4. Automated Generic Intelligence

179. 179. What is 'AI governance'?

     1. The frameworks, policies, and regulations established to guide the responsible development and use of AI ✓
     2. A type of computer hardware management system
     3. A type of network cable standard
     4. A type of file compression technique

180. 180. Why are GPUs often preferred over standard CPUs for training deep learning models?

     1. GPUs can perform many parallel calculations simultaneously, which suits the matrix operations common in deep learning ✓
     2. GPUs are always cheaper than CPUs
     3. GPUs cannot perform any mathematical calculations
     4. GPUs are only used for playing video games and nothing else

181. 181. What is 'Narrow AI' (also called 'Weak AI')?

     1. AI designed and trained to perform a specific, limited task ✓
     2. AI with human-level general intelligence across all domains
     3. AI that surpasses human intelligence in every field
     4. A type of computer hardware

182. 182. What is the field of 'swarm robotics' concerned with?

     1. Coordinating groups of simple robots to work together, often inspired by the collective behavior of social insects ✓
     2. Building a single, extremely large robot
     3. A type of computer hardware component
     4. A type of network protocol

183. 183. What is one reason organizations might choose to adopt AI-powered automation in customer support?

     1. To handle routine inquiries efficiently and provide faster responses, especially outside of business hours ✓
     2. To completely eliminate the need for any customer support entirely
     3. To increase the cost of every customer interaction significantly
     4. To make responses slower and less accurate on purpose

184. 184. At which conference was the term 'Artificial Intelligence' first formally introduced?

     1. The Dartmouth Conference ✓
     2. The Turing Conference
     3. The Stanford Symposium
     4. The MIT Summit

185. 185. What is the purpose of splitting data into 'training' and 'testing' sets in machine learning?

     1. To train the model on one portion of the data and evaluate its performance on a separate, unseen portion ✓
     2. To make the dataset smaller for no particular reason
     3. To permanently delete part of the data
     4. To increase the file size of the dataset

186. 186. What is a common feature of AI-powered 'autocorrect' on smartphones?

     1. It uses learned language patterns to automatically fix spelling errors as a user types ✓
     2. It physically prints corrected documents
     3. It only works without any internet or processing power
     4. It permanently deletes all misspelled words from the device

187. 187. Which company developed the ChatGPT conversational AI system?

     1. OpenAI ✓
     2. Microsoft
     3. Apple
     4. IBM

188. 188. What is 'natural language generation' (NLG), a subfield of NLP?

     1. The process by which AI systems produce human-readable text from data or other input ✓
     2. The process of deleting text from a document
     3. A type of computer hardware
     4. A type of network cable

189. 189. What is a 'dataset bias' most likely to result from?

     1. Training data that does not adequately represent the diversity of real-world scenarios or populations ✓
     2. Using too much computing power during training
     3. Using a very fast internet connection
     4. Using too many GPUs during training

190. 190. What does it mean for an AI system to be described as 'trained on a large corpus of text'?

     1. The system learned patterns of language by processing a very large collection of written text ✓
     2. The system was physically assembled using text-based manuals only
     3. The system requires no data to function at all
     4. The system can only process a single sentence at a time

191. 191. What is 'explainable AI' (XAI)?

     1. AI systems designed so that their decisions and reasoning can be understood by humans ✓
     2. AI systems that cannot be explained under any circumstances
     3. A type of computer hardware
     4. A type of network protocol

192. 192. Which AI program, developed by DeepMind, famously defeated world Go champion Lee Sedol in 2016?

     1. AlphaGo ✓
     2. Deep Blue
     3. Watson
     4. Siri

193. 193. What is 'overfitting' in machine learning?

     1. When a model learns the training data too closely, including its noise, and performs poorly on new, unseen data ✓
     2. When a model performs equally well on all datasets
     3. When a model has too little data to learn from
     4. When a model is too simple to be useful

194. 194. What does 'NLP' stand for in the context of AI?

     1. Natural Language Processing ✓
     2. Network Language Protocol
     3. Neural Learning Process
     4. New Language Program

195. 195. What is a 'use case' in the context of AI applications?

     1. A specific scenario or example describing how a technology can be applied to solve a particular problem ✓
     2. A type of computer hardware component
     3. A type of network cable
     4. A type of file compression format

196. 196. What is 'few-shot learning' in the context of AI models?

     1. A model's ability to perform a new task after being shown only a small number of examples ✓
     2. A model that requires millions of examples before it can perform any task
     3. A type of computer hardware
     4. A type of network protocol

197. 197. In supervised learning, what does 'labeled data' mean?

     1. Data that includes both the input and the correct/desired output ✓
     2. Data with no information attached
     3. Data that is encrypted
     4. Data stored only in the cloud

198. 198. What is a 'neural network' inspired by?

     1. The structure and function of the human brain's interconnected neurons ✓
     2. The structure of a computer's hard disk
     3. The structure of the internet
     4. The structure of a keyboard

199. 199. Which early AI program, developed in the 1960s, is considered one of the first chatbots and simulated a psychotherapist?

     1. ELIZA ✓
     2. Siri
     3. ChatGPT
     4. Watson

200. 200. What does 'RNN' commonly stand for in the context of deep learning?

     1. Recurrent Neural Network ✓
     2. Random Neural Network
     3. Rapid Neural Network
     4. Reversed Neural Network