Seven in ten US teens use AI for schoolwork, survey finds
A Common Sense Media survey of 1,017 US teens found 70% use AI for schoolwork, while only 30% have had a teacher discuss safe use.
Photo by https://kaboompics.com/ on Pexels
Seventy per cent of American teenagers now use AI tools for schoolwork, but only 30 per cent say a teacher has ever discussed with them how to use those tools safely. That is the central gap in “Teens in the AI Era: Schoolwork and Skills That Matter”, a nationally representative survey of 1,017 teenagers aged 13 to 17 that Common Sense Media published on 18 August 2026. The report is the second national dataset the organisation has released this year on teens using AI for schoolwork, and it lands as US schools reopen for the 2026-27 academic year.
The survey was fielded between 30 April and 14 May 2026 by NORC at the University of Chicago, drawing 682 probability-sample interviews from the AmeriSpeak Teen Omnibus panel and 335 non-probability interviews from Prodege, weighted to the 2026 Current Population Survey. The margin of error is plus or minus 4.3 percentage points at the 95 per cent confidence level. Authors are Laura Zimmermann, Supreet Mann, Jenny Radesky and Michael B. Robb. Most of the behavioural figures below are based on the 665 respondents who reported using AI for schoolwork, not on the full sample.
What teens using AI for schoolwork actually do with it
The report separates two patterns that overlap in the same students. Seventy-seven per cent of AI users say they use it for help without getting direct answers: 45 per cent for brainstorming or getting started, 43 per cent to check an answer after completing it themselves, and 40 per cent for feedback after writing their own version first. At the same time, 63 per cent say they get an answer to their schoolwork from AI in at least one of three ways.
| How AI users get answers from the tool | Share of teens who use AI for schoolwork |
|---|---|
| Get an answer and fix or improve it based on what I know | 35% |
| Get an answer and rewrite it to sound like me | 31% |
| Get an answer from AI and use it as-is | 25% |
| Net: get an answer in at least one of these ways | 63% |
Source: Common Sense Media, “Teens in the AI Era: Schoolwork and Skills That Matter”, August 2026. Base: 665 teens who use AI for schoolwork. The three rows are multi-select options and do not sum to the net.
These are not sequential stages. A student can brainstorm with a chatbot on one assignment and paste an answer into another, and the same respondents appear in both the 77 per cent and the 63 per cent. The report also records that 44 per cent of AI users have had a tool blocked or banned on a school network or school-issued device, and that 59 per cent of those students used a personal device to reach it anyway.
Where teens go first when they are stuck is a separate question, and the answer is not AI. Sixty-five per cent said they first try to solve the problem themselves or ask another person: 26 per cent ask a teacher or parent, 25 per cent work it out alone and 14 per cent ask a friend. Ten per cent open an AI app or tool first.
Teachers are talking about rules more than about safety
The survey asked which AI topics a teacher had ever discussed in class. Permission rules came top; verification and safety came well behind.
Common Sense Media’s own research landing page states a lower figure, 27 per cent, for teachers discussing safe AI use. The full report gives 30 per cent in Figure 6, with 27 per cent applying instead to teachers explaining what AI is and how it works. The 30 per cent figure is used here because it comes from the report itself.
School rules are also unclear to a large minority. Thirty-five per cent of teens either do not know the rules, have heard rules exist without knowing them, or say they are allowed to use AI but do not understand the conditions. Twenty-three per cent report AI is banned at their school, 33 per cent say it is allowed and they understand the terms, and 9 per cent say there are no apparent rules. Only 52 per cent think most of their teachers apply similar rules.
Alex Kotran, chief executive of the AI Education Project, told The 74 that the safety gap was the most concerning finding: “The fact that 70% of kids haven’t had a conversation about AI safety is a pretty big alarm bell, because it means the kids are using this stuff, and a lot of them are flying blind.”
Students report both benefit and loss
The report asked AI users to agree or disagree with a set of statements about their own learning. The responses point in opposite directions at similar magnitudes.
Asked which three skills will matter most in their future, teens ranked working well with other people first at 55 per cent, fact checking and critical thinking second at 50 per cent, and reading and understanding complex information third at 48 per cent. Knowing how to use AI tools effectively came sixth at 21 per cent, below writing clearly at 27 per cent.
| Skill teens expect to matter most | Share selecting it (pick three) |
|---|---|
| Working well with other people | 55% |
| Fact checking and critical thinking | 50% |
| Reading and understanding complex information | 48% |
| Science, engineering and maths | 38% |
| Writing clearly | 27% |
| Knowing how to use AI tools effectively | 21% |
| Artistic skills | 15% |
Source: Common Sense Media, August 2026. Base: 1,017 US teens aged 13 to 17.
The report notes one further split. Twenty-two per cent of the sample say they have an Individualised Education Program or a 504 plan, and 44 per cent of those students customise AI tools for their learning needs, against 34 per cent of students without one.
What the report recommends
The discussion section argues against two responses. On blame, the authors write that “we should be cautious about narratives that instinctively blame teens for using AI in schoolwork, and instead reflect on the larger culture and incentives that may underlie teens’ drive to have AI double-check everything.” On restriction, they write that “blanket bans and one-size-fits-all restrictions on AI, however well intentioned, risk cutting off tools that some students could benefit from,” and that schools should leave room for accommodations.
The authors also call for a move away from general-purpose chatbots, saying “a shift from broad, multi-purpose AI tools (which most teens in this survey reported using) to more purpose-built and rigorously tested tools will be essential.” Sixty-eight per cent of AI users in the sample reported using multi-purpose assistants such as ChatGPT, Gemini or Grok; 14 per cent used a dedicated tutor.
Yvette Renteria, chief program officer at Common Sense Media, framed the classroom problem in the organisation’s companion article: “Policies, rules, and bans – none of it matters unless you have conversations around them as an entire community.” Asked which skill matters most in the AI era, she answered: “Human connection.”
John King Jr., chancellor of the State University of New York, former US Secretary of Education and an advisory board member of Common Sense Media’s Youth AI Safety Institute, told The 74 that assessment design would have to change: “Some of that is old-fashioned blue books and saying, ‘We’re going to do some of the writing in class,’ complementing a written assignment with an oral presentation in class so you know that students are really doing the work.”
The findings echo earlier survey work covered on this site. Oxford University Press found that teenagers are split on AI in schoolwork, with only 15 per cent of UK pupils saying they had enough guidance, while the OECD and the European Commission responded to the same gap by publishing an AI literacy framework for schools.
About Common Sense Media
Common Sense Media is a 501(c)(3) nonprofit founded in 2003 by James P. Steyer and based in San Francisco. It rates media and technology products for children, publishes classroom curricula, and advocates on child online safety policy. It reports reaching more than 150 million users globally, over 1.4 million educators and more than 100,000 schools each year. Its research programme has produced repeated national datasets on youth technology use, including “The Dawn of the AI Era” in 2024 and “Unpacking Grind Culture in American Teens” the same year. In May 2026 the organisation launched the Youth AI Safety Institute, a testing and standards body modelled on car crash-test ratings, with funders including the Walton Family Foundation alongside industry contributors Anthropic, the OpenAI Foundation and Pinterest; its advisory board includes John King Jr. and Jenny Radesky. On 8 June 2026 the Institute published “The Common Sense Media Census: AI Use by Tweens and Teens”, a survey of 1,204 children aged 9 to 17 which found that 86 per cent of that age group use or interact with AI. The August 2026 survey narrows that population to teenagers and to schoolwork specifically, and arrives alongside a new Common Sense Digital Literacy and Well-Being Curriculum and an “AI and Student Well-Being” professional development course for teachers.
Sources: Common Sense Media; Common Sense Media; The 74; Common Sense Media
Featured image: photo by https://kaboompics.com/ on Pexels (free Pexels license).
Become a Sponsor
Our website is the heart of the mission of WINSS – it’s where we share updates, publish research, highlight community impact, and connect with supporters around the world. To keep this essential platform running, updated, and accessible, we rely on the generosity of you, who believe in our work.
We offer the option to sponsor monthly, or just once choosing the amount of your choice. If you run a company, please contact us via info@winssolutions.org.
I specialize in sustainability education, curriculum co-creation, and early-stage project strategy. At WINSS, I craft articles on sustainability, transformative AI, and related topics. When I’m not writing, you’ll find me chasing the perfect sushi roll, exploring cities around the globe, or unwinding with my dog Puffy — the world’s most loyal sidekick.
