News

Nearly 9 in 10 children now use AI. Their parents are the last to find out.

6 min readCommon Sense Media

Ask a room of parents how many of their children use artificial intelligence and you will get a scatter of cautious guesses. The real answer, from the first annual census of its kind, is almost all of them. And the second finding is the uncomfortable one: on nearly every measure, children report doing more with AI than their parents believe they do.

The headline number: 86%

Common Sense Media surveyed 1,204 children aged 9 to 17 for its 2026 census of AI use, the first in a series it plans to repeat annually. It found that 86% of children aged 9 to 17 use or interact with AI. Roughly a quarter do so every day, and 52% at least once a week.

The age breakdown matters more than the headline, because it shows how early this now starts:

  • 9 to 12-year-olds: 81% use AI, including 17% every single day. This is primary-school age.
  • 13 to 15-year-olds: 89%, including 28% daily.
  • 16 to 17-year-olds: 92%, including 30% daily.

There is no meaningful "not yet" group left. By the time a child is twelve, AI is not an upcoming conversation, it is an established habit.

Parents are consistently one step behind

Three separate studies published in 2026 found the same gap, measured different ways.

  • Pew Research Center surveyed 1,458 US teens aged 13 to 17 alongside their parents. 64% of teens said they use AI chatbots. Only 51% of parents thought their own child did. A further 30% of parents simply were not sure.
  • The Family Online Safety Institute, in research fielded by Ipsos across more than 4,000 US and Australian parents and children and published on 9 July 2026, found children reporting higher activity than parents recognised across almost every category. The widest gap of all was generative AI: 27% of parents said their child used it, against 38% of the children themselves.
  • FOSI also found parental optimism about AI's safety role falling from 52% to 42% in a single measurement cycle. Parents are growing more worried and less informed at the same time.

The silence behind the gap

The most actionable statistic in the census is not about usage at all. More than four in ten children said no parent or guardian has ever talked to them about AI safety. Pew found the mirror image from the adult side: only about 51% of parents had discussed chatbots with their teen, meaning roughly 40% never had.

That silence is where the risk actually lives. Children are not hiding AI use because it feels shameful. In most homes it has simply never come up, so there is no established way to raise it when something goes wrong. Common Sense also found that more than half of young AI users have asked it about their health or their body, and over one in three have used it to talk through feelings or personal problems. Those are exactly the conversations that need an adult somewhere nearby.

What good looks like at home

  • Assume use, then ask about it. Starting from "do you use AI" invites a shrug. "Which one do you use, and what is it actually good for?" gets you a demonstration and a real conversation.
  • Have the accuracy conversation once, properly. Ask it something you already know well and find the mistake together. Children who have seen it be confidently wrong once calibrate faster than children who have been warned ten times.
  • Separate the uses. Most parents are comfortable with information and schoolwork, and uncomfortable with emotional support. Say which is which out loud, because your child cannot guess your line.
  • Start younger than feels natural. At 81% usage among 9 to 12-year-olds, waiting until secondary school means arriving years after the habit formed.
  • Revisit it. This is not a one-time talk. The tools changed substantially in the twelve months these surveys covered, and they will change again.

Where SafeNeat fits

The gap these studies measure is an information gap, not a control gap, and it is worth being precise about which one a parental control app can close:

  • App visibility answers the factual question these surveys show parents getting wrong: which AI tools are actually on the device, and how much they are used.
  • Age-appropriate rules let the settings for a 9-year-old differ from those for a 16-year-old, which the 81% to 92% usage spread says they should.
  • Web filtering and time limits shape where and when, without pretending to supervise the content of every conversation.

What no app can do is have the conversation for you, and the census is clear that the missing conversation is the bigger problem. SafeNeat is privacy-first and end-to-end encrypted precisely so that closing your information gap does not mean opening your child's private life to anyone else.

The takeaway for parents

Nearly nine in ten children are already using AI, a quarter of them daily, starting at nine years old. On every survey that asked both sides, parents guessed low. And more than four in ten children have never had a single conversation with an adult about using it safely.

You do not need to become an expert on the technology. You need to stop being the last person in the house to know it is being used. That gap closes with one question asked this week, not with a setting.

Sources

  • Common Sense Media Census: AI Use by Tweens and Teens, 2026: commonsensemedia.org
  • Pew Research Center, What parents say about their teen's AI use, 24 February 2026: pewresearch.org
  • Family Online Safety Institute, Beyond Borders: U.S. and Australian Families on Online Safety, 9 July 2026: fosi.org

Want the right limits without the daily battles?

SafeNeat sets the right limits automatically, based on your child’s age.

Try it free