Teens and AI homework: where parents draw the line
Most US teens now use AI chatbots for schoolwork — here is where parents should draw the line, and the one house rule that keeps AI a tutor, not a ghostwriter.
Parents should draw the line between AI as a tutor and AI as a ghostwriter: most US teens now use chatbots for schoolwork, teens widely accept AI for research and explanations but not for writing essays, and nearly half call getting direct answers cheating — the house rule that works is 'AI explains, you write.'
Your teenager finishes a two-page history essay in four minutes flat, leans back, and says it was easy. Your stomach drops. Did they write it, or did a chatbot write it for them?
Fresh research from Pew Research Center — published in February 2026, based on a survey of 1,458 US teens and their parents conducted from September 25 to October 9, 2025 (GovTech's write-up confirms the fieldwork details) — shows that AI chatbots have quietly become a homework staple. More than half of teens have used them for schoolwork. And most parents do not realise how far in their kid already is.
This article breaks down the numbers, explains where teens and parents disagree about cheating, and gives you three house rules that turn AI from a ghostwriter back into a tutor.
How far in is your kid already? The numbers
The Pew survey is the freshest large-scale picture we have, and it is a bracing one:
- 64% of US teens say they have used AI chatbots like ChatGPT, Copilot or Character.ai, and about three in ten use them daily.
- 54% have used chatbots for help with schoolwork, making it the number-two use after searching for information (57%) and ahead of fun and entertainment (47%).
- About four in ten have used chatbots to summarise articles, books or videos.
Depth varies widely. One in ten teens say they do all or most of their schoolwork with chatbot help; larger shares say some (21%) or a little (23%); and 45% have not used chatbots for school at all.
What are they doing with it? About four in ten or more report using chatbots to research a topic or solve maths problems, while 35% have used them to edit something they wrote — a breakdown confirmed by education analyst Med Kharbach. And they find it useful: roughly a quarter say chatbots were extremely or very helpful, another quarter somewhat helpful, and only 3% say little or no help. That usefulness is why the habit is sticking — Digital Trends' coverage of the survey notes adoption keeps climbing, and a RAND survey of US students found AI use for homework rose from 48% in May 2025 to 62% by December 2025, driven by middle and high schoolers.
The cheating question: nobody agrees where the line is
Here is where it gets uncomfortable. A majority of teens — 59% — say using AI to cheat is a regular occurrence at their school, happening at least somewhat often, with about a third saying it happens extremely or very often. Only 14% say it rarely or never happens (eWeek's summary of the Pew data highlights just how normalised this feels to kids).
Yet what counts as "cheating" is genuinely fuzzy. Older Pew research shows the spectrum of teen judgement: 54% said it was acceptable to use ChatGPT to research new topics, 29% for solving maths problems, but only 18% for writing essays — and 42% said AI essay-writing is not acceptable. A more recent RAND survey (December 2025, 1,214 US students) found the same fault line: 45% called getting direct answers to homework "cheating", but nearly 80% said using AI to understand an assignment was not cheating, and most said the same for brainstorming (72%) and looking up facts (67%).
In other words: teens largely agree that AI doing the work crosses the line, and AI helping them understand does not. The argument is not about AI — it is about who did the thinking.
Tutor or ghostwriter: where to draw the line
Think of the line the way schools are starting to: a calculator is a tool in maths class and cheating in a mental-arithmetic test. AI is the same — the same app is a tutor in one context and a ghostwriter in another.
| Homework job | Tutor mode — helps learning | Ghostwriter mode — crosses the line |
|---|---|---|
| Researching | Asking for a reading list, then reading it yourself | Copy-pasting an AI summary into the assignment |
| Understanding | "Explain mitosis like I'm 14" until it clicks | Asking for the paragraph to hand in |
| Maths | Working through one example, then solving the rest yourself | Copying answers for the whole problem set |
| Writing | Getting feedback on your own draft | Pasting AI-written paragraphs into your essay |
| Studying | Turning notes into a quiz you answer yourself | Asking for the completed worksheet |
| Revision | Summarising a chapter, then writing notes from memory | Submitting the AI summary as your work |
The rule that fits every row: AI explains, you write.
The parent blind spot
Here is the uncomfortable part for parents. While 64% of teens report using chatbots, only about half of parents say their teen uses them — and close to three in ten parents are not sure whether their teen uses them at all. Most parents simply do not know how far in their kid is, a gap the Middle Earth youth counselling summary of the Pew data flags as the first thing to fix.
Schools are not much clearer: only about one-third of US students said their school had a schoolwide AI policy (RAND, December 2025). And among UK undergraduates, 12% admit pasting AI-generated text directly into assessed work — but 61% have used AI to explain concepts. Even at university level, explaining beats pasting.
The new-calculator rule: three house rules that work
Forbidding AI outright will not survive contact with a teenager's homework. Here are three rules drawn from what the research says works:
1. Ask "show me how you use it", not "are you cheating?"
An accusation ends the conversation; curiosity starts it. Ask your teen to show you exactly what they asked the chatbot and what they did with the answer. You will learn more in five minutes than a month of spot-checking essays ever taught you.
2. Make the rule "AI explains, you write."
One sentence, no exceptions. The chatbot may explain the topic, quiz them and give feedback on their draft — but the final words and answers are theirs.
3. Verify and cite.
Chatbots make confident mistakes — inventing facts, faking citations, miscalculating. Teach your teen to treat the chatbot like Wikipedia: a starting point you check, not an authority you quote.
The conversation to have before report-card season
There is a deeper point worth making to your teen. Homework is not really about the worksheet — it is about practising thinking about thinking: planning an answer, noticing confusion, re-reading, correcting. Copy-paste skips all of it.
Teens sense this themselves: about a third of teens who saw AI's impact as negative pointed to overreliance and loss of critical thinking, and 67% of students agreed heavy AI use harms critical thinking (RAND). Your teenager probably already suspects the shortcut has a cost — your job is to say it out loud before the grades do.
Have the conversation now, in October, not in a panic when the report card lands. Open with curiosity, agree the "AI explains, you write" rule together, and put it in writing. In a world where two-thirds of teens already use AI chatbots, the parents who set the rules are the ones whose kids get the tutor and skip the ghostwriter.
Frequently asked questions
Is using ChatGPT for homework cheating?
It depends on how. 45% of US students call getting direct answers "cheating", while nearly 80% say using AI to understand an assignment is not (RAND, December 2025). AI that explains is a tutor; AI that writes the work is a ghostwriter.
How many teenagers really use AI for schoolwork?
Per Pew's February 2026 survey, 54% of US teens aged 13 to 17 have used AI chatbots for schoolwork — one in ten do all or most of their schoolwork that way. Use is growing fast: RAND found it rose from 48% to 62% during 2025.
Does AI help kids learn, or make them lazier?
Teens say it helps — about half call chatbots extremely, very or somewhat helpful (Pew). But two-thirds of students agree heavy AI use harms critical thinking (RAND). The mode decides: AI that explains and quizzes builds learning; AI that pastes answers skips it.
I caught my child pasting an AI answer. What should I do?
Open with curiosity, not accusation. Ask them to show you what they asked the chatbot, then redo the task in "tutor mode": the AI explains one example, they solve the rest. Teens already accept the boundary — only 18% think AI essay-writing is fine — they just need it made explicit.
Should I just ban AI chatbots at home?
A ban is a losing battle: 95% of teens have heard of chatbots, nearly two-thirds use them, and schools are moving towards AI literacy, not prohibition. Use the "new calculator" rule instead: ban ghostwriter mode, not the tool — AI explains, your teen writes.