When AI Writes Buggy Code — How to Read It Critically

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-coding-reading-buggy-ai-code-builders

1. What is a 'fabricated API' in the context of AI-generated code?

   1. A function call that looks correct but does not actually exist in the library being used
   2. A method for automatically generating documentation for code
   3. A security feature that prevents unauthorized access to code functions
   4. A debugging tool that tracks API usage patterns

2. Why might AI-generated code fail even when it compiles without errors?

   1. Compilation errors only occur when the code is too complex for the computer to handle
   2. Code that compiles is guaranteed to produce correct results in all cases
   3. The code may use syntax from a library version that does not match what you have installed
   4. The compiler automatically fixes all logical errors before generating the executable

3. Which of these is an example of an 'off-by-one' error?

   1. A loop that never stops running
   2. A loop that processes all items except the first or last one
   3. A variable that contains the value undefined
   4. A function that returns the wrong data type

4. What are 'edge cases' that AI-generated code often misses?

   1. The variable names used throughout the program
   2. The final lines of code in a function
   3. Unusual or extreme inputs like empty strings, zero values, or null references
   4. Comments that explain complex logic

5. What is a 'silent type mismatch' in AI-generated code?

   1. When code runs without producing any output
   2. When two variables have the same name but different purposes
   3. When the code expects a number but receives a string, and the error is not immediately obvious
   4. When a function returns nothing but the program expects a value

6. According to the four-question review process, which question should you ask first?

   1. What does the code do if every input is wrong?
   2. What does the code do if every input is empty?
   3. Does every function call exist in the real library?
   4. Is there a test that proves this works?

7. Why should you copy AI-generated code into an isolated file before running it?

   1. To prevent a bad suggestion from corrupting your main project
   2. To ensure the code uses the most recent library versions
   3. To automatically generate documentation for the code
   4. To make the code run faster than in your main codebase

8. When you catch a bug in AI-generated code, what is the best approach to get it fixed?

   1. Delete the code and ask the AI to rewrite everything from scratch
   2. Tell the AI exactly what went wrong, paste the error, and ask it to fix only that specific issue
   3. Report the bug to the company that made the AI tool
   4. Ask the AI to write a completely different solution to the problem

9. The lesson compares AI-generated code to code from 'a very fast, very confident intern.' What does this comparison suggest you should do?

   1. Only read the code without testing it since interns write clean code
   2. Read it carefully, test it thoroughly, and challenge its assumptions before trusting it
   3. Replace your existing code with whatever the AI suggests
   4. Automatically trust the code since interns are usually well-trained

10. If you see an import statement for a library you have never used before, what should you do?

    1. Use it immediately since the AI clearly knows what it's doing
    2. Search for it on the package website to verify it actually exists
    3. Replace it with a more popular library that does similar things
    4. Delete it since imports are not necessary for most programs

11. What kind of inputs should you test after getting AI code to 'work' initially?

    1. Only inputs that match the expected data format exactly
    2. Only the exact inputs shown in the AI's example
    3. Weird or unusual inputs that the code might not have considered
    4. Inputs that are guaranteed to succeed

12. What is the primary purpose of 'code review' when working with AI-generated code?

    1. To compare the AI's code style with your own coding style
    2. To catch errors and verify correctness before using the code in a project
    3. To make the code look more professional by adding comments
    4. To automatically optimize the code for better performance

13. An AI generates a function that works perfectly for normal inputs but crashes when given an empty list. What kind of bug is this?

    1. A missing edge case
    2. A fabricated API
    3. An off-by-one error
    4. A syntax error

14. After the AI fixes a bug you reported, what should you do next?

    1. Delete the original code since the bug is now fixed
    2. Share the fix with other developers immediately
    3. Re-run the code to confirm the fix actually works
    4. Ask the AI to explain why the bug happened in detail

15. What makes AI-generated code particularly dangerous compared to human-written code?

    1. It requires much more testing than human-written code
    2. It always uses the most outdated programming patterns
    3. It looks completely convincing and smooth while containing hidden errors
    4. It cannot be run on standard computer systems