When AI Checks Its Own Homework

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-explorers-agentic-AI-and-the-helper-that-checks-itself-r9a5

1. What is a self-checking AI agent?

   1. An AI that never makes mistakes
   2. An AI that reads and evaluates its own answers before sharing them
   3. An AI that only checks other people's work
   4. An AI that checks the weather

2. What happens when a self-checking AI writes code that crashes?

   1. It gives up immediately
   2. It runs the code, reads the error, and tries to fix the bug
   3. It blames the user
   4. It deletes the code

3. Why do self-checking agents make fewer mistakes?

   1. They are programmed to never be wrong
   2. They review their own work and catch errors before presenting the result
   3. They use better computers
   4. They work slower so mistakes don't happen

4. What does it mean to 'verify' something in AI?

   1. To make it run faster
   2. To delete it if it's wrong
   3. To check that it is correct or works as expected
   4. To copy it to another computer

5. When an AI writes a poem and reads it back to find something to improve, what is it doing?

   1. Deleting the poem
   2. Grading other students' work
   3. Self-checking its own creative output
   4. Asking the user for help

6. Why is self-checking compared to proofreading homework?

   1. Both involve copying someone else's work
   2. Both involve reviewing your own output to catch errors before it's submitted
   3. Both require a teacher to help
   4. Both make you work slower permanently

7. What should you ask an AI to trigger self-checking?

   1. Tell me a joke
   2. Review your own answer and point out any mistakes or things you're unsure about
   3. Work faster please
   4. Delete your last response

8. If an AI grades its own writing, what is it demonstrating?

   1. Laziness
   2. Self-checking ability
   3. Overconfidence
   4. A broken program

9. What do agents do when they find mistakes in their own work during self-checking?

   1. They immediately shut down
   2. They pretend the mistakes didn't happen
   3. They try to fix the mistakes before presenting the result
   4. They charge extra money

10. What makes an AI an 'agent' rather than just a simple program?

    1. It costs more money
    2. It can take multiple actions in sequence to complete a task
    3. It runs on a special computer
    4. It was made in a different country

11. What makes a self-checking agent better than one that shows its first answer immediately?

    1. Self-checking agents are always right
    2. Self-checking agents catch errors before the user sees them
    3. Self-checking agents work offline
    4. Self-checking agents cost less

12. Why might a self-checking AI that tests its own code be better than one that doesn't?

    1. Because it works louder
    2. Because it catches bugs before the user runs the code
    3. Because it costs more
    4. Because it only writes short programs

13. What happens when a self-checking AI tests its code and it breaks?

    1. It gives up and asks you to fix it
    2. It shows you the broken code and says 'good luck'
    3. It reads the error message and tries to fix the code itself
    4. It turns off the computer

14. What is 'iteration' in the context of a self-checking AI?

    1. Giving up after one try
    2. Repeating a cycle of doing, checking, and improving until the result is good
    3. Showing the same answer three times
    4. A type of computer processor

15. You ask an AI to write a poem, and it says 'Here's a draft — I'd change the rhythm in line 3.' What is this AI demonstrating?

    1. Rudeness to the user
    2. Self-checking: it reviewed its own output and found something to improve
    3. A broken program
    4. Asking for more money