Skip to main content

AI Ethics & Responsible Use

Reading time: ~8 minutes

Who Wrote This Code?​

When AI generates code for you, who is the author?

You are. The moment you paste AI-generated code into your project, you take full responsibility for it. If it has a bug, that's your bug. If it has a security vulnerability, that's your vulnerability.

This is true in school, at work, and in this hackathon.

Think of AI like a colleague who gives you advice. You can listen, but the decision — and the responsibility — is yours.

The Copy-Paste Trap​

The biggest risk with AI coding tools is the copy-paste trap:

  1. You ask AI to generate code
  2. It looks reasonable
  3. You paste it into your project without really reading it
  4. It works (or seems to)
  5. You move on

The problem: You don't understand what the code does. When something breaks later, you can't debug it. When you need to modify it, you don't know how. You've built your project on code you don't own mentally.

How to Avoid the Trap​

Follow the RUMT approach — for every piece of AI-generated code:

StepWhat to doWhy
ReadRead every line of the generated codeMake sure you can follow the logic
UnderstandCan you explain what it does to someone else?If not, ask the AI to explain it
ModifyChange at least one thing to fit your specific needsThis forces you to engage with the code
TestRun it and verify it works correctlyDon't assume — check

If you can't complete all four steps, you shouldn't use that code.

When NOT to Trust AI​

AI coding tools are impressive, but they have real limitations. Be especially careful in these areas:

1. Security​

AI-generated code often has security vulnerabilities:

  • SQL injection (user input not sanitized)
  • Cross-site scripting (XSS)
  • Hardcoded passwords or API keys
  • Insecure authentication logic

Rule: Never trust AI for security-critical code without careful review.

2. Edge Cases​

AI generates code that works for the "happy path" (normal conditions) but may fail on:

  • Empty inputs
  • Very large numbers
  • Special characters
  • Unexpected data types
  • Network failures

Rule: Always think about what could go wrong.

3. Business Logic​

AI doesn't know your specific business rules. It might generate a discount calculation that's technically correct but uses the wrong discount percentage, or a date function that doesn't account for your company's fiscal year.

Rule: You define the rules. AI implements them. Always verify the rules are correct.

4. Performance​

AI-generated code often works but is not optimized. It might:

  • Use inefficient algorithms (O(n²) when O(n) is possible)
  • Make unnecessary API calls
  • Load more data than needed
  • Not handle caching

Rule: For performance-critical code, review and optimize.

5. Outdated Information​

AI models are trained on data up to a certain date. They might suggest:

  • Deprecated functions or libraries
  • Old API syntax that has changed
  • Outdated best practices

Rule: Check the official documentation for current syntax.

AI Ethics During the Hackathon​

During the hackathon, keep these principles in mind:

Do:​

  • Use AI tools freely — that's the whole point of this research
  • Ask AI for help when you're stuck
  • Use AI to learn new concepts and techniques
  • Discuss AI suggestions with your team before using them
  • Review all AI-generated code before adding it to your project

Don't:​

  • Copy-paste code you don't understand
  • Let AI do all the work while you watch
  • Ignore errors or warnings in AI-generated code
  • Use AI to bypass thinking — use it to enhance thinking

Remember:​

The hackathon is not a competition. There are no grades. The goal is to observe how you work with AI tools naturally. Be yourself — use AI the way you normally would (or want to learn to).

The Bigger Picture​

AI coding tools are not going away. They will only get more powerful. The question is not whether to use them, but how to use them responsibly.

The developers who thrive in the AI era will be those who:

  • Can evaluate AI output critically
  • Understand the code they use, regardless of who (or what) wrote it
  • Know when AI is helpful and when it's a risk
  • Communicate effectively with both AI and human teammates
  • Take responsibility for the final product

These are exactly the skills this research is studying.

Next: Tutorial 6: Setup — Codeflow Recorder (video)