Here is a quiet gap in many American homes. In a Pew Research Center survey of 1,458 US teens and their parents, published in February 2026, 64% of teens said they use AI chatbots. About three in ten use them daily. But only 51% of the parents in the same survey said their teen uses one, and another 28% were not sure.
The same families, two different answers. That gap is the real story, and it is the reason this guide exists.
This guide is not about banning or panicking. It is about closing the gap with one good conversation. It is general information, not medical or mental health advice.
The Surprise: It Is Mostly Homework
Before worrying, look at what teens say they actually do. According to the same Pew survey, among teens who use chatbots:
- 57% use them to search for information
- 54% use them for schoolwork help
- 47% use them for fun
- 42% use them to summarize articles, books or videos
- 16% use them for casual conversation
- 12% use them for emotional support or advice
So for most teens, a chatbot works like a fast, chatty search engine and homework helper. That is good news, and it means the first conversation does not need to be a confrontation.
It does not mean there is nothing to watch. Ten percent of teens in the survey said they do all or most of their schoolwork with chatbot help, and 59% said students at their school use AI to cheat at least somewhat often. Parents were comfortable with searching for information (79%), but far fewer were comfortable with a teen using a chatbot for emotional support (18%).
That last number points to where your attention is best spent.
The One Category That Deserves Extra Attention
There is a difference between a chatbot used as a tool and a chatbot used as a friend.
Some apps are built as “AI companions,” meaning characters designed to chat, flatter and act like a friend or partner. A 2025 survey of 1,060 teens by Common Sense Media found that 72% had used AI companions, and about half used them at least a few times a month. A third of teen users (34%) said something a bot said or did made them uncomfortable, and 25% had shared real personal information such as their name, location or secrets.
Common Sense Media rated companion apps an “unacceptable risk” for young users. To be fair, most of those teens also said they spend more time with real friends than with the bots, and many found the bot conversations less satisfying than human ones. This is not a story of every teen being pulled in. It is a story of a risk worth knowing about.
Why the concern? A chatbot is a prediction machine built to sound warm and agreeable. Feeling understood by it is not the same as being understood, and the long-term effects on young people are still being studied.
Five Questions to Ask Your Teen This Week
Experts quoted in a PolitiFact guide agree on one thing: ask directly, stay curious, and do not lead with anger. Psychologist Mitch Prinstein of the American Psychological Association suggests telling your teen you will not be angry about what you find, only that you care.
You do not need to ask all five at once. Try one at the dinner table or in the car.
- “What do you use AI for?” Let them show off. Ask them to teach you something.
- “What is the best thing it has helped you with?” This keeps the conversation friendly.
- “Has it ever told you something that turned out to be wrong?” This builds the habit of checking answers.
- “What do you think it knows about you?” This opens the door to privacy.
- “Has anything it said ever made you uncomfortable?” Say this one calmly, and mean it. If the answer is yes, thank them for telling you.
An illustrative example: a mother in Texas asks her 15-year-old to show her how she uses AI for chemistry. The daughter proudly explains that it quizzes her before tests. Two days later the mom asks, “Do you ever just chat with it about other stuff?” Because the first talk went well, the second is easy.
Make a Family AI Plan (One Page Is Enough)
Researcher Naomi Aguiar of Oregon State University recommends a written plan created together, instead of rules handed down. Sit down and fill in a simple page:
- School rule. AI may explain, quiz and give feedback. It may not write what the teacher asked you to write.
- Privacy rule. No full name, address, school name, photos of other people, or passwords typed into a chatbot.
- Companion rule. No apps designed as friends, partners or therapists.
- Honesty rule. If something it says feels strange, scary or too personal, tell a trusted adult. No one gets in trouble for telling.
- People first. For big feelings, we talk to a person first.
- Check-in. A short family check-in once a week, with no lecture.
Psychiatry fellow Akanksha Dadlani of Stanford emphasizes supervision, transparency and boundaries over complete prevention. Model it yourself: let your teen see you double-check an AI answer.
Warning Signs Worth Noticing
Experts cited by PolitiFact list the following as signs of unhealthy use. Any single one can be normal teenage behavior, so look for patterns.
- Pulling away from friends, family or activities
- More secrecy, or long stretches alone with devices
- Distress when access is limited
- Treating a chatbot like a therapist or best friend
- Sudden changes in grades, sleep or eating
If a teen seems to rely on a chatbot over people while withdrawing from others, psychotherapist Grace Berman advises involving a mental health professional.
Common Problems and Practical Solutions
Problem: “My teen says everyone uses it, so why can’t I?” Solution: Agree with the true part. Most do use it. Then focus on how, not whether. “How” is a conversation. “Whether” becomes a fight.
Problem: I suspect AI is doing the homework. Solution: Ask your teen to explain the answer out loud. If they can, the AI helped them learn. If they cannot, it did the work for them. Our guide on how teachers use AI in the classroom shows what schools consider acceptable.
Problem: My teen only talks to the chatbot about feelings. Solution: Stay calm and do not take the tool away in anger. Ask what they like about it. Gently offer other outlets, such as a school counselor, a doctor, or a family member. If it keeps up, speak with a professional.
Problem: I do not understand the technology myself. Solution: Start with AI in daily life, which shows how much of it is already in your family’s apps.
Problem: My teen got a message that felt like a scam. Solution: Teens are targets too. See our guides to AI scams and deepfakes and digital fraud protection.
Problem: I want to use parental controls. Solution: Check the settings of each app your teen uses, and turn off long-term “memory” features where you can. Prinstein notes that controls work only when both the parent and the teen accounts are set up and linked, so involve your teen in setting them up.
If You Are Worried About Your Teen’s Safety
If your teen has talked about harming themselves, or you believe they are in danger, call or text 988 (the Suicide and Crisis Lifeline) in the US, or call 911 in an emergency. A chatbot is not a crisis service, and neither is a parent’s search history. Reach out to people.
Where AI Is Good for Teens
Used well, it can quiz a student before a test or explain a hard idea three ways. Our list of best AI tools for students and the broader piece on artificial intelligence in education cover the positive side. Our post on AI in social media also helps explain where much of teens’ AI contact begins. For more simple guides, you can learn more about us on the AI Overview Search about page.
Frequently Asked Questions
How many teens use AI chatbots? A Pew Research Center survey published in February 2026 found that 64% of US teens aged 13 to 17 use them, and about three in ten use them daily.
Are AI companion apps safe for kids? Common Sense Media has called them an “unacceptable risk” for young users. Experts generally advise avoiding apps marketed as friends, partners or therapists.
Should I ban AI chatbots? Most experts quoted here favor supervision, open conversation and clear family rules over a complete ban. Your own decision depends on your teen’s age and needs.
Is it okay if my teen uses AI for homework? Using AI to explain, quiz and give feedback is generally fine. Having it write the assignment is not, and may break school rules. Check your school’s policy.
What if my teen shared personal information with a chatbot? Stay calm. Help them delete the conversation where possible, change any passwords they shared, and talk about what is safe to type next time.
