Lesson 1 Quiz — Core AI/ML Vocabulary

A retail company wants a model to respond to a customer's request the instant it comes in — no delay, no batching. Which type of inferencing is this?

A company has years of purchase history with no predefined categories, and wants the system to discover natural customer segments on its own. Which learning type is this?

Which statement correctly distinguishes an 'algorithm' from a 'model'?

A model scores 99% accuracy on training data but performs poorly on new, unseen data. What is this called?

Which best explains why Agentic AI is shown as a separate box connected to GenAI, rather than nested inside it?

A company collects sensor readings from IoT devices every second, and the order of readings matters for detecting anomalies. What data type is this primarily?

A system learns to play a strategy game via trial and error, getting reward/penalty feedback, with no dataset of correct answers upfront. Which learning type is this?

A customer submits a request for a long report; because it takes minutes to generate, they don't wait live — they're notified later. Which inferencing type is this?

A company runs a job every night scoring 2 million transactions for fraud risk, in one large process, nobody waiting on any individual result. Which inferencing type is this?

A startup wants to run occasional, unpredictable inference requests without managing servers, paying only for compute used. Which inferencing type best fits?

A hiring model trained mostly on resumes from one demographic group systematically favors that group even with equal qualifications. What is this called?

A bank has 50,000 loan applications each labeled 'approved'/'denied', and wants to train a model to predict approvals for new applications. Which learning type is this?

A model performs poorly on both training data and new data — it never captured the underlying pattern. What is this called?

A dataset of raw video files with no labels — what type of data is this?

Which correctly orders these from broadest to narrowest, based on nesting?

A factory uses cameras and a trained model to spot defective products on an assembly line. Which AI capability is this?

A chatbot needs to understand sentiment and meaning behind a customer's typed message and generate a text reply. Which AI capability is this?

Which best distinguishes 'fairness' from 'bias'?

Which best describes gradient boosting compared to linear regression?

A folder of images each tagged with what object it contains ('cat', 'dog') — what type of data is this?

A logistics company predicts delivery delays using GPS pings every 10 seconds (each historical trip labeled 'delayed'/'on-time'), alerting drivers instantly. Which combination of (1) data type, (2) learning type, (3) inferencing type is correct?

A foundation model is deployed behind an API that scales to zero when idle (serverless), and separately a nightly job scores 5 million records using that same model. Which statement is correct?