Lesson 128 Quiz: AI Costs and Token Usage

Question 1

Which type of content contributes to input token usage in an AI request?

Only the final answer displayed to the user Only the application name and database name User messages, instructions, and retrieved context Only the users login information

Question 2

Why can sending a long conversation history increase AI usage?

Previous messages may become part of the input context Conversation history automatically reduces token usage Previous messages are converted into database indexes Long conversations prevent all model processing

Question 3

Which approach can reduce unnecessary AI requests when the same stable result is needed repeatedly?

Adding more conversation history every time Caching reusable results Increasing output length for every request Repeating the same request multiple times

Question 4

Why can an AI agent cost more than a simple single-response AI application?

Agents always use smaller models than simple applications Agents do not use AI models after the first step Agents may perform multiple model and tool operations Agents have no measurable usage cost

Question 5

Which combination helps control unexpected AI usage and cost in production?

Long prompts and unlimited requests only Rate limits, usage monitoring, and budget controls Removing all application monitoring Allowing every user unlimited AI requests