Table of Contents
How to Talk to Kids About AI Safety: 44% Never Have
How to talk to kids about AI safety when 44% say no parent ever has: what kids say they need, plus a first-conversation script for ages 5 to 8, 9 to 12 and 13+.
Forty-four percent. That’s the share of US kids aged 9 to 17 who told Common Sense Media this spring that no parent or guardian had ever talked to them about how to use AI safely. Among kids who don’t use AI yet, it’s 64%.
If you’re figuring out how to talk to kids about AI, that second number is the useful one. The conversation is easiest before the habit exists, and most families are skipping that window. This guide covers what the surveys say kids actually need to hear, why parents put the talk off, and a first-conversation script for each age that takes about ten minutes.
Key Takeaways
- In Common Sense’s June 2026 census, 56% of kids said a parent had talked to them about AI safety and 44% said no one had.
- Pew found just over half of parents of teens talk with them about chatbots, and lower-income parents were less likely to (43% vs 56%).
- Kids who saw something inappropriate from an AI mostly kept quiet: only 33% told a trusted adult.
- Only 51% of kids had heard from a school or teacher how to tell whether AI information is accurate.
- The best first conversation is a question, not a lecture: “Show me how you use it.”
The 44% Gap: What the Surveys Found
The AI safety conversation is the talk in which a parent explains how to use AI tools without being harmed by them: what to share, what to trust, and what to do when something feels wrong. Two large 2026 surveys measured how often it happens.
Common Sense Media’s June 2026 census of 1,204 kids, fielded March 18 to 26 by SSRS, found:
- 56% of kids say a parent or guardian has talked to them about using AI safely; 44% say no one has.
- Among kids who already use AI, 59% have had the talk. Among kids who don’t, only 36%.
- By age: 53% of 9- to 12-year-olds and 57% of 13- to 17-year-olds.
- Schools do somewhat better on rules: 73% of kids say a teacher has explained what they should and shouldn’t use AI for in schoolwork. But only 56% have heard about safety, and 51% about how to tell if AI information is accurate.
Pew Research Center’s survey of parents of teens, fielded in fall 2025, found “just over half” of parents say they talk with their teen about chatbots, while roughly four in ten haven’t. Parents in households earning $75,000 or more were more likely than those earning under $30,000 to have talked (56% vs 43%). And 51% of parents said their teen uses chatbots, against 64% of teens who said so themselves.
Why the Talk Gets Skipped
Parents don’t skip it out of indifference. A few patterns stand out in the data.
The kid isn’t using AI yet, so it feels early. The census shows the gap is widest exactly there. But use climbs fast: Common Sense’s October 2026 census found 36% of 9-year-olds and 73% of 13-year-olds have used a chatbot.
Parents don’t know their teen uses it. Pew’s 13-point gap between what teens report and what parents believe means some parents think there’s nothing to discuss.
Parents don’t use it themselves. It’s hard to explain a tool you’ve never opened. That’s a fixable problem, covered in our 20-minute catch-up for parents who don’t use chatbots.
Safety talks feel like they need expertise. They don’t. The core content fits on an index card.
What Kids Say They Need
The best guide to the conversation comes from kids themselves.
They need permission to report. Among chatbot users in the June census, 17% said an AI had shown or told them something that “wasn’t OK or wasn’t meant for someone their age.” Only 33% of those kids told a trusted adult; 53% didn’t. A safety talk that makes reporting feel safe closes most of that gap.
They need accuracy skills. Only half had heard from school how to check AI answers.
They want limits, not bans. When about 100 students from all 50 states drafted a student AI bill of rights in July 2026, The 74 reported that they approved mandatory AI literacy covering misinformation, plagiarism, bias and privacy by 83 to 15, and called AI useful for brainstorming and studying but not for test answers.
And when they’re struggling, they need a person. The JED Foundation’s 2026 parent tip sheet, based on a survey of more than 5,500 students, advises parents to “stay close and curious” and to ask open questions such as “Do you ever talk to AI when you’re stressed or anxious?”
How to Teach Your Kid About AI Safety, Age by Age
This is the first conversation, not the only one. Keep it short, do it next to a screen, and end with something your child can repeat back.
Ages 5–8: “It’s a computer that guesses words”
When: The first time your child hears a voice assistant or sees you use a chatbot.
Say: “This is a computer program. It doesn’t think or feel. It guesses what words come next, and sometimes it guesses wrong. We only use it together.”
Try: Ask it a question your child knows the answer to, like what color their favorite stuffed animal is, and laugh together when it can’t know.
End with: “If a computer or game ever says something weird or scary, you tell me, and you won’t be in trouble.”
Ages 9–12: “Show me, then let’s set three rules”
When: Now. This is the age when the census shows use climbing steeply.
Ask first: “Have you or your friends used anything like ChatGPT or an AI in a game? Show me.” Let them drive. Don’t correct yet.
Then say: “AI is useful, and it makes mistakes. It sounds sure even when it’s wrong. And what you type gets saved by the company.”
Agree on three rules, written down: (1) no names, school, address or photos of yourself; (2) AI can explain homework, but you write the answer; (3) if it says something not okay, you tell me.
End with: Ask them to explain the rules back in their own words.
Ages 13+: “What’s it good for, and when isn’t it?”
When: Any calm moment, ideally not right after a problem.
Ask: “What do you use AI for? What’s it actually good at? Has it ever been wrong in a way that mattered?” Teens often have sharp opinions. In a Gallup survey for the Walton Family Foundation, 80% of people aged 14 to 29 said relying on AI to work faster will likely make learning harder.
Then add the two things teens most need to hear: a chatbot can sound caring but isn’t a friend or a counselor, and it often agrees with you even when it shouldn’t. If they’re going through something hard, a person can help in ways a bot can’t.
Close with resources, casually: if they or a friend are ever in crisis, call or text 988.
The question to ask: “If an AI told you something that made you uncomfortable, what would you do next, and who would you tell?”
How to Talk to Kids About AI: Choosing Your Opening
| Your situation | Best opener | What to avoid |
|---|---|---|
| Child doesn’t use AI yet | ”This is what AI is, and here’s how we’ll use it together” | Waiting until they start |
| Child uses AI for homework | ”Show me how you use it for school” | Opening with cheating accusations |
| Teen uses AI more than you do | ”Teach me. What’s it good at?” | Pretending to be the expert |
| You don’t use AI at all | ”Let’s try it together for ten minutes” | Banning what you haven’t seen |
| Child saw something upsetting | ”Thank you for telling me. You’re not in trouble.” | Taking the device away on the spot |
Making the First Talk Stick
Repeat it in small doses
One conversation won’t hold. Bring it up when AI is in the news, when a new app appears, or at the start of a school term. Our guide to running a ten-minute weekly family tech talk gives a format.
Explore together
The JED tip sheet suggests exploring AI tools together to build trust. Ask your child to show you something they made or learned with AI. Curiosity gets you more information than rules do.
Pair the talk with settings
Talk first, then review privacy and parental settings side by side, so your child knows why each one exists.
Revisit what AI is, not just the rules
For the bigger picture of explaining AI without anxiety, see our earlier guide on how to talk to your child about AI without overwhelming them.
What not to do
Don’t make the first conversation the day you discover a problem. A talk that starts as a punishment teaches kids to hide the next problem. And don’t promise surveillance you won’t keep; teens notice.
What to Watch For Over the Next 3 Months
- Week 4: Your child can state the family’s AI rules in their own words.
- Month 2 red flags: Your child becomes secretive about an AI app, says a chatbot “gets them” better than people, or stops mentioning AI at all after previously talking about it.
- Month 3 self-check: Has your child told you about at least one AI answer that was wrong or weird? That’s the sign the reporting channel works.
Frequently Asked Questions
At what age should I talk to my child about AI safety?
By around 9, and earlier if they use voice assistants or AI toys. Common Sense’s 2026 census found 36% of 9-year-olds have already used a chatbot, and the safety conversation gap is widest among kids who haven’t started yet, which is the easiest time to set norms.
What should I tell my kid about AI?
Three things: AI is a computer program that predicts words and can be wrong while sounding sure; what you type is saved by the company, so keep personal details out; and if it ever says something that feels wrong, tell an adult without fear of getting in trouble.
How do I talk to my teen about AI if they know more than I do?
Ask them to teach you. Have them show you what they use it for, then ask what it’s good at and when it fails. Teens are often skeptical of AI already. Your job is adding the parts they may not weigh: privacy, emotional reliance, and the habit of checking.
What if my child already saw something inappropriate from an AI?
Thank them for telling you, stay calm and make clear they’re not in trouble. Ask what happened, report it in the app if possible, and review settings together. If the content involved self-harm or left your child distressed, talk with their doctor or call or text 988.
Does my child’s school already cover AI safety?
Partly. In Common Sense’s census, 73% of kids said a teacher explained what AI is allowed for in schoolwork, but only 56% heard about using it safely and 51% about checking whether AI information is accurate. Home conversations fill that gap.
About the author
Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.
Sources
- Common Sense Media. (2026). “The Common Sense Media Census: AI Use by Tweens and Teens, 2026.” Common Sense Media. https://www.commonsensemedia.org/sites/default/files/research/report/2026-ai-use-by-tweens-and-teens-1.pdf
- Pew Research Center. (2026, February 24). “What parents say about their teen’s AI use.” Pew Research Center. https://www.pewresearch.org/internet/2026/02/24/what-parents-say-about-their-teens-ai-use/
- Mann, S. (2026, October 27). “Almost half of tweens have used an AI chatbot or companion, new research finds.” Common Sense Media Youth AI Safety Institute. https://institute.commonsensemedia.org/tween-teen-census-2026
- Toppo, G. (2026, July 24). “Students Convene to Hammer Out AI Bill of Rights for Schools.” The 74. https://www.the74million.org/article/students-convene-to-hammer-out-ai-bill-of-rights-for-schools/
- The Jed Foundation. (2026). “AI and Your Teen: What Parents and Caregivers Need to Know.” JED. https://assets.jedfoundation.org/pdfs/AI_and_Mental_Health_Three-Pager_Digital_2026.pdf
- Walton Family Foundation. (2026, April 9). “Gen Z Resentment Toward AI Grows as Adoption Stagnates and Workplace Fears Mount.” Walton Family Foundation. https://www.waltonfamilyfoundation.org/about-us/newsroom/gen-z-resentment-toward-ai-grows-as-adoption-stagnates-and-workplace-fears-mount
