Problem Sets, Code, and Group Projects
Module 3 · AI in Your Assignments

Problem Sets, Code, and Group Projects

Use AI to get unstuck on math and science problems, debug code you wrote, and work fairly in group projects, without letting it do the work you're there to learn.

Problem Sets: Get Unstuck, Not Carried

Math, science, and other problem-based courses build skill through practice. Each problem you solve yourself makes the next one easier. AI can solve most homework problems instantly, which makes it easy to “finish” homework and then freeze on the test.

Try first, then ask for the smallest help that gets you moving:

Instead ofTry
”Solve this problem.""Here’s my work so far. What’s the next step? Don’t solve it."
"Show me the answer.""I got 42 but the answer key says 36. Where did I go wrong?"
"Explain how to do this type of problem.""Give me a similar problem with different numbers so I can practice.”
I'm working on this physics problem: [paste]. Here's what I've done
so far: [your work]. I'm stuck. Give me one hint about what to think
about next. Don't give the next step or the answer.

Discernment for math and science: AI makes calculation and logic errors, sometimes with total confidence. Check its steps, and remember your answer key and teacher are the authority for your course.


Code: Understand What You Submit

In programming courses, AI coding assistants are powerful and the rules vary a lot, so check your course policy carefully. When AI help is allowed:

  • Write it first. Your attempt, even broken, is where learning happens.
  • Ask AI to explain errors, not just fix them: “What does this error message mean, and why is it happening?”
  • Never submit code you can’t explain. If you couldn’t walk your teacher through every line, you don’t understand it yet, and it will show on exams and interviews.
  • Test everything. AI-generated code can look right and still fail on edge cases.
Here's my function and the error I'm getting: [paste]. Explain what
the error means and give me a hint about where the bug is. Don't
rewrite my code.

Group Projects: Agree Before You Start

Group projects add a new problem: everyone’s grade depends on everyone’s AI choices. One teammate’s undisclosed AI use can put the whole group at risk. Settle it at the first meeting:

  1. What does the course allow for this project? Everyone reads the policy.
  2. How will each of us use AI, if at all?
  3. How will we check AI-assisted parts for accuracy?
  4. How will we disclose it in the final submission?

Also: don’t paste teammates’ work or messages into AI without asking them. Their work is theirs to share.

⚠️ Common Pitfall: The Teammate Who Used AI Secretly

Common mistake: discovering the night before the deadline that one section was AI-written against the rules, with fake sources.

Agree on AI use at the start, and review each other’s sections, including sources, before submitting. If something violates the policy, fix it before submission or talk to your teacher. Don’t hope nobody notices.

🧪 Try It: The Smallest-Hint Challenge
  1. Find a problem from a current class that you can’t immediately solve (math, science, coding, or logic).
  2. Work on it yourself for at least five minutes. Write down where you got stuck.
  3. Ask AI for one hint only, using the prompt above.
  4. Try again. Need more help? Ask for one more hint. Count how many hints you needed.
  5. Once you’ve solved it, ask AI for a similar problem with different numbers, and solve it with no hints.
  6. Reflect: Could you solve a third one on a test tomorrow?
🧠 Knowledge Check

1. You're stuck halfway through a math problem. What's the best AI request?

2. What's the most important rule for using AI in a programming course where it's allowed?

3. When should a group agree on how it will use AI?


Up Next

Module 4 covers honesty: how to disclose AI use, what to do if you’re ever accused of misusing AI, and building AI skills for your future.