Parent Guides

How to tell if your child is using AI for homework (and what to do about it)

8 min read · Updated October 10, 2026

You usually cannot tell from the finished homework alone, and AI detectors are not reliable enough to settle it. The better signs are a gap between the work and what your child can explain, a sudden change in their writing voice, and answers that skip the steps their class teaches. The best response is a calm conversation, the school’s own AI rules, and setting AI up as a tutor rather than an answer machine.

AI help with homework is now ordinary. In a Pew Research Center survey of US teens aged 13 to 17, carried out in the fall of 2025, more than half said they had used AI chatbots to get help with schoolwork, and about one in ten said they do all or most of their schoolwork with a chatbot’s help. So for most families the useful question is not whether your child uses AI for school. It is how they use it, and whether that use is helping them learn or doing the learning for them.

When AI homework help is fine, and when it crosses a line

It helps to think of AI use for school as a range rather than a yes or no. At one end, AI works like a patient tutor. At the other, it does the assignment and your child hands it in. Most of the uses in between depend on the class rules.

• Usually fine: asking for a concept to be explained again in simpler words, generating practice questions before a test, or getting feedback on a draft they wrote themselves.
• Depends on the class: using AI to outline an essay, to check grammar, or to brainstorm ideas for a project.
• Usually not allowed: having AI write an essay, a lab report or short answers that are then handed in as their own work, or copying worked solutions to graded problems.

Teens know this is a live issue. In the same Pew survey, a majority said they think students at their school regularly use AI to cheat. That does not mean your child does. It does mean the pressure, and the easy option, are there for almost every student.

Signs your child may be leaning on AI for homework

None of these signs proves anything on its own. Children’s writing improves, some children genuinely work fast, and some are simply having a good week. Look for a pattern, and treat what you notice as a reason to ask, not as a verdict.

• The work does not sound like them. The vocabulary, tone or polish is noticeably different from their messages, their earlier assignments or how they talk about the subject.
• They cannot explain it. Ask them to walk you through one paragraph or one problem. Work they did themselves is usually easy to talk about, even if it is not perfect.
• Right answers with missing working. Maths answers appear without the steps, or use a method their class has not covered yet.
• A gap between homework and tests. Homework grades rise while quiz and test results, done in class without AI, stay the same or fall.
• Confident but slightly wrong details. AI tools can produce quotes, sources or facts that sound right and are not, such as a book quotation that does not appear in the book.
• Homework that used to take an evening now takes ten minutes, with nothing else about the subject having changed.
• A teacher mentions that their written work and in-class work do not match.

Why AI detectors will not settle it

It is tempting to paste an essay into an AI detector, or to ask ChatGPT whether it wrote something. Neither gives you an answer you can rely on.

OpenAI, the company that makes ChatGPT, says plainly in its own Help Center that, in its experience, AI detectors do not work reliably. When it tried to build one, the detector labeled human writing, including Shakespeare and the Declaration of Independence, as AI-generated. OpenAI also says detectors may unfairly flag students learning English and students whose writing is formulaic or concise, and that small edits can get AI text past them.

Asking a chatbot does not help either. OpenAI says ChatGPT has no knowledge of what content could be AI-generated, and its answers to "did you write this?" are random. A false accusation built on a detector score can do real damage to your relationship with your child, so leave detection out of it.

What to do if you think AI is doing the homework

1. Start curious, not accusing. "Show me how you did this" opens a conversation. "Did you cheat?" usually ends one. How to talk to your teen about AI has more openings that do not start an argument.
2. Ask them to walk you through the work. If they can explain it, the question may be answered. If they cannot, ask how they used AI, and listen to the whole answer before reacting.
3. Find out the rule for that class. Check the syllabus or the school’s AI policy, or email the teacher. Rules differ between teachers, and your child may honestly not know where the line is.
4. Look for the reason. A child who hands homework to AI is often stuck, overloaded or worried about grades. Fixing that, with time, help or a conversation with the teacher, does more than any rule.
5. Agree on a family line together, such as "AI can explain, quiz and give feedback, but the words and the working you hand in are yours." Family AI rules has a set of rules you can adapt.
6. Set AI up to teach rather than answer, using the learning modes below.
7. If AI-written work has already been handed in, encourage your child to speak to the teacher themselves, with your support. Owning it early usually goes better than being found out.

What to say

Turn AI into a tutor instead of an answer machine

The major AI services now include modes built for learning. They ask questions and work through a problem with your child instead of handing over the answer.

• ChatGPT study mode is available on every ChatGPT plan. Instead of only giving a final answer, it asks questions, explains ideas step by step and checks understanding as it goes. OpenAI says it can still sometimes give a direct answer, so it is a help rather than a guarantee.
• Study hours, for ChatGPT teen accounts linked to a parent. A parent can choose times when new conversations start in study mode, for example homework hours on school nights. Study hours do not block ChatGPT. Parental controls for ChatGPT shows how to link accounts and find the setting.
• Gemini Guided Learning, launched by Google in August 2025, breaks problems down step by step and asks open-ended questions rather than jumping straight to the answer. Gemini parental controls covers the account side. For other AI apps, parental controls in every AI app lists what each one offers.

These modes work best when your child understands why you are suggesting them. A child who sees study mode as a way to actually do well on the test, not as a punishment, is far more likely to keep using it. If your child uses ChatGPT, ChatGPT safety for teens covers the wider picture beyond homework, and AI chatbot age limits explains which services allow their age at all.

Where TeenCue fits

TeenCue gives parents a view of AI use without reading over their child’s shoulder. It runs in Chrome, Firefox, Edge and Brave on your child’s computer and describes a chat on supported AI websites with a category such as schoolwork, a more specific topic, and a concern rating when something may be worth a look. You can see, for example, whether most of their AI use is schoolwork and which subjects it is about, which is often the easiest way into the conversations above.

It does not show full conversations, so it cannot tell you whether a particular essay was written by AI, and it does not block AI services. It does not cover phones or tablets. It is designed to be used with your child’s knowledge. How TeenCue works lists the services it currently supports.