How Generative AI Creates Content - Quiz

Question 1

What is the basic workflow of a Generative AI system?

Prompt or Input → Model Processing → Generation → Output Output → Training → Prompt → Database Database → Output → Training → Prompt Prompt → Delete Model → Output → Training

Question 2

What is a prompt?

An input or instruction provided to guide the AI system A physical component of a computer The final generated answer A copy of the training dataset

Question 3

How does a language model generally generate a long text response?

It generally generates the response sequentially by predicting tokens It generates the entire response as one physical object It always retrieves an exact paragraph from a database It must retrain the entire model after every token

Question 4

What can generation settings such as temperature influence?

They can influence how predictable or varied the generated output is They permanently change the training dataset They guarantee factual accuracy They determine the physical size of the computer

Question 5

What is the difference between a prompt, a model and an output?

The prompt is the input, the model processes it, and the output is the generated result The prompt and output are always the same thing The model is the final answer shown to the user The output is the training dataset

Question 6

Can Generative AI generate content other than text?

Yes, including images, audio, video and code No, Generative AI can only produce text No, Generative AI can only classify images No, Generative AI can only process numerical data

Question 7

Does generating fluent content guarantee that the result is correct?

No. Generated content can contain errors even when it sounds convincing Yes. Fluent language guarantees factual accuracy Yes. Generative AI automatically verifies every statement Yes. A model cannot generate incorrect code

Question 8

Why can external information be useful to a Generative AI application?

It can provide current or specific information that the model can use during generation It permanently replaces the trained model It guarantees every generated response is correct It eliminates the need for any model training