How to Talk to Your Kids About AI and Homework (by Age, With Scripts)

Many parents find themselves somewhere between two instincts. One says: ban it, they need to learn the old way. The other says: it's the future, they need to learn to use it. Both are partly right, and neither is a plan.

What children need is what they've always needed with a powerful new tool: an adult who explains what it's for, what it's not for, and why. This guide gives you the conversation for each age, the words to use and a simple family agreement.

Before you start: three principles

Curiosity before rules. Ask what your child already uses and how. You'll probably learn something. Children who feel interrogated hide things. Children who feel listened to tell you.

Reasons before rules. "Because I said so" loses to "but everyone uses it" every time. "Because homework is how your brain practices" is much harder to argue with.

The goal is learning, not avoiding AI. Your child will use AI throughout their education and working life. The skill you're teaching is using it in a way that makes them smarter, not in a way that replaces their thinking.

Ages 5 to 8: "It's a helper, not a know-it-all"

Young children don't need a lecture on large language models. They need a simple mental model and a few habits.

What to explain:

• A computer can talk, but it's not a person, and it doesn't really know them.
• It can be wrong, even when it sounds sure.
• We don't tell computers private things, like our address or school.
• A grown-up is always nearby when they use it.

Words you can use:

"This is a computer program that's good at talking. It's not a person, and sometimes it gets things wrong. If it ever says something weird or confusing, come and tell me. You won't be in trouble."

At this age, the tool matters most. Children this young should only use AI designed for them, with a narrow purpose and a parent close by. General chatbots aren't built for this age group, and their own terms usually require users to be at least 13.

Ages 9 to 12: "It should help you think, not think for you"

This is the age when homework gets harder and the temptation to shortcut appears. It's also when children can understand why the shortcut costs them.

What to explain:

• Homework is practice. If the AI does the practice, the AI gets better at math and they don't.
• There's a difference between asking for help and asking for answers.
• AI can make mistakes in math that are hard to spot.
• Private information stays private: full name, school, address, photos of themselves.

Words you can use:

"Think of homework like soccer practice. If someone else runs the drills for you, you'll still be slow on game day. AI can be like a coach who gives tips, or like someone who runs the drills for you. I want it to be the coach."

A good test to teach: after any help, "could I do one like this by myself?" If not, the help wasn't the right kind. We explain this in more detail in is using AI for homework cheating?

Ages 13 and up: "You're in charge of how you use it"

Teenagers are likely already using general chatbots, often more fluently than their parents. Rules imposed from above tend to be quietly ignored. Rules worked out together tend to stick.

What to explain:

• The research: students who used a chatbot for answers during practice did better that day and worse on the test afterwards. (A 2025 study in PNAS found a 17% drop.)
• Schools have policies, and the consequences of breaking them can be serious. Know your teachers' rules.
• AI can be confidently wrong. They're responsible for what they submit.
• Chatbots are designed to be engaging. Notice if it's replacing people or sleep.

Words you can use:

"I'm not going to ban it. You'll use AI for the rest of your life and I want you to be good at it. Being good at it means using it to get smarter, not to skip the hard parts. What do you think a fair rule would be?"

Let them set some of the rules. Teenagers often propose stricter rules than you would, and they're far more likely to follow rules they wrote.

When the conversation goes wrong

"Everyone uses it." "You're probably right. I'm not saying don't use it. I'm saying use it in a way that helps you on the test, when it's not there."

"It's just like a calculator." "Sort of. A calculator does the arithmetic so you can think about the problem. If AI does the thinking too, what's left for you to learn?"

"The teacher will never know." "Maybe not on the homework. But the test will know."

Silence or eye-rolling. Let it go for now and come back to it. One calm conversation is better than one long one.

A family AI agreement

Some families find it helps to write the rules down. Here's a starting point you can adapt:

1. I try first. I attempt every problem before I ask AI for help.
2. I ask for hints, not answers. My first question is "what should I do first?"
3. I can explain it. If I can't explain how I got an answer, I haven't finished.
4. I'm honest about it. If AI helped me, I can say so.
5. I keep private things private. No full name, address, school or photos of myself.
6. I check its work. AI can be wrong. I estimate and plug answers back in.
7. I tell a grown-up if AI says something strange, upsetting or confusing.
8. I follow my school's rules.

Review it every few months. What makes sense at 10 won't at 14.

Make the right choice the easy one

Rules are easier to follow when the tool is on the same side. If the only AI available is a general chatbot that answers everything, your child is working against it every time they try to follow rule two.

PennPaper is designed to follow these rules by default. The tutor sticks to math. It asks your child to try each step first, gives hints rather than final answers, and explains ideas on a whiteboard as it talks. There's no chatting with strangers, and you can read what happened in every session from the parent dashboard. It's a way to say "yes, use AI" and mean the right kind.

For more on choosing a tool, see how to choose a safe AI tutor for kids, and if you're worried about general chatbots specifically, is ChatGPT safe for kids?

Frequently asked questions

At what age should I talk to my child about AI?

As soon as they encounter it, which for many children is before age 8. Start with simple ideas: it's a computer, it can be wrong, and we keep private things private.

Should I ban AI for homework?

Bans are hard to enforce and tend to push use out of sight. Clear rules, reasons and the right tool usually work better.

How do I know if my child is following the rules?

Ask them to explain one problem from each assignment. If they can, the learning happened. We list other signs in signs your child is using AI for homework.

What if I don't understand AI myself?

You don't need to. The conversation is about learning and honesty, which you already understand. Try the tool together with your child. You'll learn alongside them.

The bottom line

The goal isn't to keep your child away from AI. It's to raise a child who uses it to think better rather than to think less. That starts with a conversation, and it's easier than you might expect.

---

Sources: Bastani et al. (2025), Generative AI without guardrails can harm learning, PNAS; OpenAI Terms of Use