Why AI Can Make Mistakes - Quiz

Question 1

Why can an AI system produce a convincing answer that is incorrect?

Because AI generates outputs without a built-in guarantee that every statement is true Because AI systems deliberately try to give false answers Because AI cannot process language Because every AI answer is randomly generated

Question 2

What is an AI hallucination?

An AI system generating incorrect or unsupported information as though it were valid A hardware failure in the computer A type of training dataset A method for improving internet speed

Question 3

Why can training data affect the reliability of an AI system?

Errors, bias, or incomplete information in training data can influence the model Training data has no relationship to model behaviour Only the size of the dataset matters Poor training data always makes a model faster

Question 4

Why can an AI system give an outdated answer?

The information available to the model may not include newer events or changes AI systems automatically forget every answer after one minute Language models cannot process dates All AI systems permanently stop learning after deployment

Question 5

Why can an ambiguous prompt cause problems?

The system may need to make assumptions about what the user means Ambiguous questions automatically retrain the model The model deletes its training data Ambiguous questions always produce correct answers

Question 6

Does fluent and confident language guarantee that an AI answer is correct?

No. Fluent language does not guarantee factual accuracy Yes. Confident wording guarantees correctness Yes. Language models cannot make factual mistakes Yes. Every fluent response has been checked by a human

Question 7

How can external retrieval systems help an AI application?

It can provide current or specific information from external sources during inference It permanently replaces the underlying model It guarantees every answer will be correct It eliminates the need for any training

Question 8

Why is verification important when an AI answer has important consequences?

Important AI-generated information can contain errors and should be checked when accuracy matters Verification is never necessary AI systems automatically verify every statement Verification is only needed when the AI response is short