Your Child Is Already Asking AI for Personal Advice. Here Is the Skill That Protects Them.

A new study stopped me cold. A Common Sense Media survey of more than 1,200 kids ages 9 to 17 found that nearly one in four children said they would turn to AI for help with homework, health questions, or personal problems before asking a teacher, counselor, or parent.^[1] Not after. Before. I have three daughters, and when I read that number, I tried to picture which of those categories applied to each of them. All three, probably. That got me thinking about a skill most parents are not explicitly teaching: knowing who to ask.

Nearly 9 in 10 kids now use AI, and a growing share turn to it for advice before talking to any adult in their lives.

The Common Sense Media Census: AI Use by Tweens and Teens, published June 8, 2026, surveyed 1,204 children ages 9 to 17 across the United States.^[1] Nearly 9 in 10 reported using or interacting with AI. One in four does so daily. The study is the first of its kind at this scale, specifically designed to track how this generation’s relationship with AI changes year over year.

What kids use AI for goes well beyond homework help. The survey found children turning to AI for entertainment, creative projects like generating images and videos, health and body questions, and advice about their future goals and decisions.^[2] The homework angle gets the most attention. But the health and personal advice piece is what stood out to me as a parent. Those are the conversations I assumed my kids would bring to me first.

There is a literacy gap underneath all of this that is easy to miss. Only about one in three children surveyed understood that AI cannot reliably distinguish truth from false information.^[1] Nearly half had never discussed AI safety with a parent or trusted adult.^[1] Kids are forming real opinions and making real decisions based on AI answers, and most of them do not understand the core limitation that makes those answers unreliable.

When kids default to AI for advice, they skip the conversations that help them develop judgment, which is the very thing they need to use AI well.

Think about what actually happens when a child asks a parent for help with something hard. The parent does not just answer. They ask follow-up questions, share context from their own experience, and sometimes say they do not know but let’s figure it out together. That back-and-forth is how kids learn to think about problems, not just solve them. When AI becomes the default, that whole layer of the interaction disappears.

The emotional dimension matters too. Ten percent of all kids in the study said they feel AI understands them better than people do.^[1] Among daily users, that number jumped to nearly one in five.^[1] I do not think that reflects something sinister about AI. I think it reflects kids who have not yet found the human connections that feel that safe. AI fills a gap when those connections are missing. That matters more than any policy conversation about AI use.

My oldest is 13. She is deeply independent, researches everything herself, and is rarely comfortable asking for help. I can easily picture her finding AI more efficient than a parent conversation that might feel slow or intrusive. That is not a parenting failure. It is a 13-year-old thing. But the instinct to turn to AI first for a health question, without knowing whether the answer is reliable, is a real and current risk for her age group.

The skill is knowing who to ask: the judgment to match a question to the right source, whether that is AI, a parent, a doctor, or a trusted friend.

Most conversations about AI and kids focus on whether kids should use it at all, or how to stop them from using it for homework. I think those are the wrong questions. The question worth sitting with is: does my kid know when AI is the right tool and when they genuinely need a human? That distinction, applied consistently, is the skill that protects them in a world where AI is everywhere.

Knowing who to ask is not the same as being skeptical of AI. AI is genuinely useful for looking up how things work, getting a starting point on a topic, or brainstorming ideas. But it is not a substitute for a doctor when your kid notices a strange symptom, a parent when a friendship is falling apart, or a counselor when something feels really wrong. Kids can learn that distinction. It just has to be taught.

The Common Sense Media data points to exactly what happens when this skill is missing. Kids ask AI for personal advice without knowing AI cannot reliably separate truth from fiction. They feel emotionally understood by a system that is pattern-matching on text. They skip the adults who actually know them. None of that is inevitable. It is a skill gap. And skill gaps can be closed.

Watch for the moments when your kid explains why they went to a person instead of a search engine, and the times they ask you something they could have easily Googled.

One of the clearest signs is when a child checks a source. If yours says something like, "I looked it up, but I wanted to ask you because I wasn’t sure if it was right," that is the skill at work. They are not avoiding AI. They are calibrating trust. They understand that an AI answer is a starting point, not an ending point. That habit, seeking a second opinion from someone with real knowledge or real experience, is exactly what knowing who to ask looks like in practice.

Another sign: your child volunteers that they came to you instead of AI because the question felt personal. My middle daughter does this naturally. She processes things out loud and comes to me not because she could not find an answer elsewhere, but because she wants to feel understood, not just informed. That instinct, knowing when the point of a conversation is connection rather than information, is a signal that she is building something AI cannot replicate. She is learning to tell the difference between a knowledge problem and a human problem.

This week, have one conversation that makes AI’s limits concrete for your kid, and open the door to bring the real questions to you.

Try the “who would know better” question. Pick something your kid might ask AI, like a health symptom, a conflict with a friend, or a big decision. Ask them: if you needed the most reliable answer on this, who would be the best source? AI, you, a doctor, a friend who has been through it? Let them work through the reasoning out loud. The goal is not the right answer. It is the habit of asking the question before defaulting.

Walk through one AI answer together and find where it falls short. Look up something your kid cares about, then compare the AI’s answer to a more reliable source. The study found that only one in three kids understands AI cannot reliably separate truth from false information.^[1] A concrete example in the moment lands harder than a general warning.

Make yourself easier to ask. Nearly half of kids have never discussed AI safety with a parent.^[1] Some of that is on us. If we have not had that conversation, kids default to whatever is most available. Open the door by asking what they use AI for, not with suspicion, but with genuine curiosity. The easier you are to approach, the less the default to AI matters.

The goal is not to compete with AI. It is not to convince your kid that asking a parent is always better. It is to give them the judgment to make that call on their own, based on the type of question, what is at stake, and who actually knows the answer.

One thing I keep coming back to is that my daughters are not doing something wrong when they go to AI first. They are doing what every generation does with new tools: using them for everything, and learning the limits over time. My job is not to make AI feel dangerous. It is to speed up the part where they learn what AI is actually good at and what it is not equipped to handle.

Knowing who to ask will be a core life skill for this generation. AI can give your kid fast information. What it cannot do is show up with experience, love, and the specific knowledge of your kid. That is still us. It just does not feel useful unless we stay available and easy to approach. That starts with a conversation this week.