Clear, practical advice on kids and screens.
You can log in any time under More.
Short answers to the questions parents actually ask. The answer first, then the reasoning. The full thinking is in the article each one links to.
30 questions
Not in itself, and I wouldn't take it away. What I'd rather avoid is it becoming the first stop when something hurts. That's when your child misses the practice of being understood by a human who isn't automatically on their side.
From the article: When your child confides in AI instead of you
It's a smaller group, often lonely or vulnerable kids. Among Norwegian children aged 9 to 16, 15 per cent use AI for health advice or to share worries. Of those who do, one in four says it understands them better than most people.
From the article: When your child confides in AI instead of you
Usually it's about how easy it is to go to. The chatbot is awake at 23:40, hasn't had a hard day, doesn't get disappointed or scared on your child's behalf, and doesn't interrupt. Those are product features, and hard for any human to match.
From the article: When your child confides in AI instead of you
Nobody knows yet. Kids who use AI daily or weekly report more loneliness than those who rarely do, but it's at least as likely that lonely children turn to the chatbot more often. A child leaning hard on it may need more people around them.
From the article: When your child confides in AI instead of you
No. These are chatbots built to be someone to talk to, and the specialists who assessed them concluded they aren't safe for anyone under 18. Character.AI has itself closed its open conversations to minors. Ordinary AI tools like ChatGPT are a different matter.
From the article: When your child confides in AI instead of you
Yes. One in six kids who have used an AI chatbot has been served inappropriate content, and only one in three of those told an adult. Worth agreeing in advance that your child shows you anything odd, you won't get angry and nobody loses anything.
From the article: When your child confides in AI instead of you
Save a standing instruction asking it to ease off the agreement and push back when it disagrees. In ChatGPT the field sits under Settings, Personalization and “Custom instructions”, and from there it covers every new conversation. It's at least as useful on your child's account.
From the article: When your child confides in AI instead of you
By lowering the threshold. Three things set it: you're available, your reaction is predictable, and honesty never costs your child anything. A good start is ten predictable minutes without the phone every day, perhaps at bedtime, when your child knows you can be reached.
From the article: When your child confides in AI instead of you
I wouldn't. It's fine for practical things and thinking out loud. Say instead that when something hurts, you'd like them to talk to you or another adult they trust. A real person who knows them can see how they're doing and be there for them.
From the article: When your child confides in AI instead of you
For most children it isn't dangerous, but as I see it, it's a short-term disservice. A chatbot that is there around the clock and nearly always takes the child's side removes the practice of being understood by someone who sees things differently.
From the article: AI is not your friend
Because agreeing was rewarded in training. Language models are trained on human feedback, and people score the answers they like highest. Stanford researchers measured it: chatbots sided with the user 49 per cent more often than people, even when the user had hurt someone.
From the article: AI is not your friend
Two ways stand out, as I read the research. A chatbot that takes the child's side leaves them surer they were right and less willing to make up after a row. One that serves up the homework answers means they learn less than without AI.
From the article: AI is not your friend
Yes, if it gives hints rather than answers. Students who practised maths with a standard GPT-4 chat did 17 per cent worse on the exam than students without AI, while a version that only gave hints seemed to remove the harm.
From the article: AI is not your friend
Link your account to your child's in ChatGPT and open the parental controls. There I'd first set “Study Mode” as the default for new conversations, so the model asks questions and gives hints rather than answers. In the research, hints seemed to remove the harm.
From the article: AI is not your friend
Give it a permanent instruction. Chatbots generally follow instructions, including one to push back when you're wrong. Save it on the account and it covers every new conversation. On your child's account, I'd do it together with your child and explain why.
From the article: AI is not your friend
I'd rather they didn't. When something hurts, we need people who know us and understand feelings in a way robots don't. In our house the rule is that the kids come to us before they ask a chatbot about anything painful.
From the article: AI is not your friend
Yes, fairly common. In an American survey, 72 per cent of teenagers had tried an AI companion, and 33 per cent had taken something serious to the chatbot instead of to a person. Specialists who assessed the apps found them unsafe for anyone under 18.
From the article: AI is not your friend
Yes, within some limits. At home AI is allowed for schoolwork as long as it isn't serving up the answers, and for anything creative. They'll grow up with it, and I'd rather they learn it with their eyes open than on the sly.
From the article: AI is not your friend
Set it on the phone itself. AI is also in Snapchat, WhatsApp and the browser, so ChatGPT's quiet hours only stop one app. On iPhone it's Downtime in Screen Time, on Android it's in Family Link, ideally from your evening cut-off through to the morning.
From the article: AI is not your friend
Ask it to, with a standing instruction it remembers. Here's the text to paste in, and where it goes in ChatGPT, Claude, Gemini and the rest.
13, and under-18s are supposed to have a parent's permission. The limit is OpenAI's own, mostly enforced by a birth date the child types in themselves. ChatGPT opening the door at 13 doesn't make 13 the right moment for your child.
From the article: Guide to kids and AI
Yes, after an honest attempt of their own. In one study, students who practised maths with a standard AI chat scored 17 per cent worse on the exam than students without AI, while those given hints did no worse. ChatGPT's “Study Mode” gives hints instead.
From the article: Guide to kids and AI
Yes. You invite your child's account from yours and set everything from there: quiet hours, “Study Mode” as default so it gives hints rather than answers, memory and model training off so conversations build no profile, and image generation and voice mode off by age.
From the article: Guide to kids and AI
No. You manage the settings, but the conversations are deliberately kept from you, and I think OpenAI got that right. Your child needs a room where embarrassing questions feel safe. If the system spots signs of serious crisis, you're contacted by email, text and phone.
From the article: Guide to kids and AI
Not without checking. AI gets things wrong in the tone it uses when it's right. Only one in three children understand it can't reliably tell fact from fabrication. Have your child quiz it on something they know inside out and see if the answer holds.
From the article: Guide to kids and AI
Because it's trained on human feedback: people rate answers they like more highly, so agreeing pays. In one study, models sided with the user about 50 per cent more often than humans. Ask it to push back and save the instruction on your child's account.
From the article: Guide to kids and AI
I'd be careful. Kids who confide in a chatbot are often lonely or vulnerable, and the AI is trained to take their side. In our house, AI is never the place to go with something that hurts. Ask your child to come to you first.
From the article: Guide to kids and AI
No. The age limit is 18, and a risk assessment by Common Sense Media and Stanford found AI companions, chatbots built to be a person, unsafe for anyone under 18. Character.AI closed its open conversations to minors. Ordinary tools like ChatGPT are a different discussion.
From the article: Guide to kids and AI
Not without Snapchat+. My AI sits at the top of the chat list, chosen by nobody, and has no age limit of its own. Easiest is to treat it as part of the app when you decide whether your child should have Snapchat at all.
From the article: Guide to kids and AI
Mostly not, going by Norway's guidance. The education authority recommends that Years 1 to 7 generally get no access, careful use in Years 8 to 10, and that upper secondary learns to use it. They're recommendations, and a useful anchor for the rules at home.
From the article: Guide to kids and AI
