Bob's AI Academy

Question 1

A manager says: "We need a system that can analyze customer inquiries and automatically suggest responses based on previous interactions." Which capability is the organization MOST likely seeking?

A) Artificial Intelligence B) Spreadsheet Automation C) Network Virtualization D) Data Archiving

Question 2

A retailer uses historical sales data to predict which products customers are likely to purchase next month. Which AI concept is being applied?

A) Computer Vision B) Machine Learning C) Robotics D) Encryption

Question 3

Which statement BEST describes the relationship between Artificial Intelligence and Machine Learning?

A) Artificial Intelligence is a subset of Machine Learning B) They are unrelated technologies C) Machine Learning is a subset of Artificial Intelligence D) They are identical concepts

Question 4

A company uses AI to generate product descriptions, marketing emails, and social media posts. Which category of AI is MOST likely being used?

A) Predictive Analytics B) Data Mining C) Expert Systems D) Generative AI

Question 5

A user asks an AI assistant: "Summarize this report in three bullet points for senior executives." What is the user's instruction commonly called?

A) Dataset B) Model C) Prompt D) Parameter

Question 6

Two employees use the same AI tool. One enters: "Write a report." The other enters: "Write a one-page executive summary for a manufacturing company focusing on operational efficiency and cost reduction." Why is the second employee more likely to receive a useful result?

A) The AI has more computing power B) The AI was retrained C) The employee used a larger language model D) The employee provided a more effective prompt

Question 7

A hospital uses AI to review medical images and highlight areas that may require physician attention. What is the MOST appropriate description of AI's role?

A) AI replaces physicians entirely B) AI assists professionals by providing recommendations C) AI guarantees accurate diagnoses D) AI eliminates the need for medical expertise

Question 8

Which scenario BEST demonstrates a real-world application of AI?

A) A calculator adding two numbers B) A printer producing documents C) A recommendation engine suggesting movies based on viewing history D) A keyboard displaying typed characters

Question 9

An AI assistant confidently provides information that later turns out to be completely incorrect. Which AI limitation is MOST likely responsible?

A) Data Compression B) Data Replication C) Virtualization D) Hallucination

Question 10

A company plans to use AI to support important business decisions. Which practice is MOST important?

A) Removing all human involvement B) Trusting all AI outputs automatically C) Avoiding monitoring after deployment D) Maintaining human review and oversight

Question 11

A machine learning team discovers that customer records contain duplicate entries, missing values, and inconsistent formatting. What is the MOST likely consequence?

A) Improved model fairness Reduced data quality that may negatively affect model performance C) Increased explainability D) Stronger governance controls

Question 12

An AI model used for hiring was trained primarily on data from one geographic region. After deployment, the organization notices lower accuracy when evaluating candidates from other regions. What is the MOST likely cause?

A) Model Encryption B) API Failure C) Sampling Bias in the training data D) Secret Management Issues

Question 13

A company wants to reduce the likelihood that an AI system produces unfair outcomes across different demographic groups. Which activity is MOST appropriate?

A) Increase cloud storage capacity B) Evaluate fairness metrics during testing and validation C) Reduce network bandwidth D) Disable monitoring systems

Question 14

A bank removes race from its loan approval model but continues using a variable that strongly correlates with race. What type of bias risk remains?

A) Measurement Bias B) Proxy Bias C) Data Drift D) Hallucination Risk

Question 15

A healthcare organization stores patient names, addresses, medical histories, and identification numbers. How should this information be classified?

A) Public Information B) Operational Metadata C) Personally Identifiable Information (PII) D) Synthetic Data

Question 16

An organization wants to analyze trends within customer data while minimizing the risk of exposing information about specific individuals. Which privacy technique is MOST relevant?

A) Data Replication B) Data Compression C) Differential Privacy D) Data Deduplication

Question 17

A company generates artificial customer records that resemble real-world patterns but do not correspond to actual individuals. What type of data is being used?

A) Structured Data B) Production Data C) Synthetic Data D) Archived Data

Question 18

A deployed fraud detection model gradually becomes less accurate because customer purchasing behavior changes over time. Which issue is MOST likely occurring?

A) Data Encryption Failure B) Model Registry Error C) Data Drift D) Access Control Failure

Question 19

An AI operations team continuously tracks latency, accuracy, error rates, and model performance after deployment. What activity are they performing?

A) Data Labeling B) Feature Engineering C) Monitoring and Observability D) Prompt Engineering

Question 20

A company uses AI to recommend financial products to customers. Even though the model performs well, all final recommendations must be reviewed by a financial advisor before action is taken. Which principle is BEST demonstrated?

A) Full Automation B) Zero Trust C) Human Oversight D) Data Minimization

Question 21

A company deploys an AI system that helps prioritize job applicants. The model performs well during testing, but leadership wants assurance that fairness, accountability, documentation, and oversight requirements are consistently followed. Which discipline is MOST responsible for addressing these concerns?

A) Data Visualization B) Network Administration C) AI Governance D) Database Management

Question 22

A security team discovers that a publicly accessible repository contains an active API key used to access a production AI service. What is the MOST appropriate action?

A) Continue monitoring usage until the next scheduled review B) Retrain the AI model immediately C) Revoke or rotate the exposed credential and investigate potential misuse D) Increase model accuracy thresholds

Question 23

An organization deploys an AI model that assists with loan approval decisions. Regulators later request evidence showing how the model was evaluated, what risks were identified, and who approved deployment. Which artifact would BEST support this request?

A) Firewall Configuration B) API Gateway Logs C) Model Documentation and Audit Evidence D) Cloud Resource Inventory

Question 24

A user repeatedly submits carefully crafted prompts designed to bypass an AI assistant's safety controls and force unintended behavior. Which threat is MOST likely being attempted?

A) Data Drift B) Membership Inference C) Prompt Injection D) Data Lineage Manipulation

Question 25

An organization adopts AI across multiple business units. Executives want a framework that helps identify risks, evaluate impacts, and implement controls throughout the AI lifecycle. Which approach BEST supports this objective?

A) Increasing model size B) Eliminating human review C) Expanding training datasets without validation D) Applying structured AI risk management practices and governance frameworks

AI Fundamentals Test