Article

New Research Finds Teens Are Turning To AI Chatbots After Something Harmful Happens Online

Home / Articles / New Research Finds Teens Are Turning To
Heard & Believed · 2026-08-01 · 6 min read

Reviewed by Heard & Believed · Updated 2026-08-01

A new survey of more than a thousand U.S. teens found that after an unwanted online sexual interaction, kids are more likely to ask an AI chatbot what to do than to tell an adult right away. Here is what that pattern says about what young people actually need next.

In short

  • A sixth annual youth safety survey from a leading child-protection nonprofit found that 67 percent of surveyed minors have used an AI chatbot or companion.
  • Among teens who reported an unwanted online sexual interaction, 15 percent said they turned to an AI chatbot for guidance afterward, roughly twice the rate of teens dealing with bullying.
  • Researchers frame this as young people processing confusing or harmful experiences with whatever tool feels closest at hand, not as a rejection of the trusted adults in their life.
  • A chatbot can offer a low-pressure first stop, but it cannot report harm, connect a teen to local resources, or notice a pattern building over time the way a person can.
BEYOND THE BOT
The Survey, By The Numbers
1,000+
U.S. minors ages 9 to 17 surveyed for this sixth annual youth online safety research
67%
of surveyed minors say they have used an AI chatbot or companion
15%
of teens who had an unwanted online sexual interaction turned to an AI chatbot for guidance afterward
2x
the rate at which those teens sought chatbot guidance, compared with teens dealing with bullying instead

Figures reflect Thorn's 2026 youth online safety survey, released July 28, 2026.

What The Survey Found

A child-safety nonprofit that has run this youth research for six consecutive years surveyed more than a thousand U.S. minors between the ages of nine and seventeen this year. Two-thirds of them, 67 percent, said they have used an AI chatbot or companion of some kind, a number that has been climbing steadily as the technology becomes ordinary rather than novel.

The more striking finding sat inside a smaller slice of that group. Among teens who said they had experienced an unwanted online sexual interaction, 15 percent reported turning to an AI chatbot for guidance afterward, about twice the rate of teens who reached for a chatbot after being bullied instead.

Why A Chatbot Might Feel Safer Than A Person First

It is not hard to understand the appeal in the moment. A chatbot is available at three in the morning, does not flinch, does not make a face, and will not immediately take away a phone or a device the way a frightened parent sometimes does. For a teen who is scared, embarrassed, or unsure whether what happened was even a big deal, that can feel like the lowest-risk place to start.

Researchers involved in this work describe it as young people using whatever tool is closest at hand to process something confusing, not as evidence that they do not trust the adults around them. It is a reflex, not a verdict on any particular relationship.

What A Chatbot Cannot Actually Do

A conversational tool can offer information and a place to think something through without immediate judgment, but it cannot file a report, verify whether a situation is ongoing or escalating, or connect a teen with a trained specialist who can help in real time. It also has no memory of a pattern building across weeks or months the way a parent, teacher, or counselor might notice.

That gap matters most in exactly the situations this survey highlights, cases involving unwanted sexual contact online, where a next step, blocking someone, saving evidence, or reporting to a platform, often needs a human being with context and authority behind it.

What It Takes To Be The Person A Teen Turns To Next

The most useful response to this research is not panic about chatbots. It is building enough trust that a teen feels safe enough to loop a person in too, even after they have already talked something through with an app. That usually means staying calm on the outside even when you are not calm on the inside, and being clear in advance about what will and will not happen if they tell you something hard.

It also helps to know, ahead of time, one real resource you can name on the spot, a hotline, a school counselor, a local advocacy center, so you are not scrambling to figure out next steps in the same moment a teen is finally ready to talk.

Ways To Be The Adult A Teen Actually Turns To

None of this requires being a tech expert. It mostly requires being predictable, calm, and specific.

  1. Ask without leading the answer: Try something open like has anything online felt confusing or upsetting lately, instead of a question that assumes the worst.
  2. React calm first, worried second: A visibly panicked reaction in the moment can quietly teach a teen to keep the next hard thing to themselves too.
  3. Say what will not happen: Telling a teen upfront that you will not immediately take their phone or ban the internet can lower the stakes of telling you anything at all.
  4. Know one real resource by name: Having a specific hotline or local advocacy center in mind ahead of time means you are not searching for help in the moment it matters most.
  5. Explain plainly what a chatbot cannot do: Gently note that an AI tool cannot report a crime or notice a pattern the way a trusted adult can, without shaming a teen for having used one.
  6. Revisit the conversation later: A teen who was not ready to talk today may be ready next week, so treat a quiet moment as an opening, not a closed door.

This is supportive information, not legal or medical advice. If you need someone now, the RAINN hotline is 800-656-4673 — free and confidential, 24/7.

You might also read

Questions

You might be wondering

Not necessarily. Researchers describe this as reflecting how comfortable this generation is with the format, not a rejection of the adults in their life. It is an opening to talk about what a chatbot can and cannot do.

The survey defined it broadly, and researchers found the reported rate climbs even higher once deepfakes and sextortion attempts are included alongside more familiar unwanted messages or images.

It can offer information and a low-pressure place to think something through, but researchers behind this survey are clear that a chatbot is not a substitute for a trusted adult or a trained support specialist.

The National Sexual Assault Hotline and child-safety organizations like the one behind this survey offer guidance built specifically for these situations, including help for parents who are unsure what to say first.

You don’t have to do this alone

Share only what feels okay. We’ll gently connect you with confidential support — no pressure, no cost.

This is a supportive resource, not legal advice, and reaching out creates no obligation.

We hear you

Someone caring will reach out within a day. If you need to talk now, RAINN is here 24/7 at 800-656-4673.