AI Chatbots and Teen Mental Health: A Parent's Guide

Quick answer

Can an AI chatbot help a struggling teenager, or does it risk making things worse? Both are possible, and the difference comes down to how it is used. Putting a hard feeling into words, even to a chatbot at 11pm when no one else is awake, can be a genuinely useful first step. What it cannot do is replace a counsellor: it cannot properly assess risk, cannot follow up, and has no way of bringing in the people who actually need to know if something serious is happening. This guide sets out what chatbots can reasonably offer a teenager, the signs that use has tipped into reliance, and the real supports - GP, school counsellor, Kids Helpline - that should stay firmly in the picture regardless of how much your teenager talks to AI.

What an AI chatbot can reasonably offer

Used lightly, a chatbot can be a low-stakes place to think out loud. A teenager who is embarrassed to say something out loud to a parent might type it to a chatbot first, and find the words easier the second time, to a real person. It can help someone organise a jumble of thoughts before a harder conversation, or simply be company at an hour when everyone else is asleep.

That is a genuinely useful function, and it is worth acknowledging rather than dismissing. The mistake is assuming it stops there.

Why a chatbot is not a counsellor, and where over-reliance creeps in

A general-purpose AI chatbot - the kind built for homework help, writing or general conversation - was not designed, trained or credentialed to manage a mental health crisis. It cannot judge how serious a situation really is, it will not notice a pattern across weeks the way a school counsellor who sees a student regularly might, and it cannot pick up the phone if things escalate. It can also simply get things wrong, offering generic or misjudged responses to something that needed a careful, human read.

This is the same caution that applies to AI companion apps, only sharper. Australia's eSafety Commissioner found that the companion apps it examined had no meaningful age checks, with some children using them for hours a day and conversations crossing into sex and self-harm. A study chatbot is a milder version of that same risk: not built for this, and not equipped to catch what a trained person would. The distinction between a companion app and a learning tool is covered in more depth in AI companion apps: what parents should know.

There is a broader pattern worth knowing, even outside mental health specifically. Gerlich's 2025 study in Societies, of 666 participants, found that heavy AI use was strongly associated with "cognitive offloading" - letting the tool do the thinking - and that this effect was strongest in 17- to 25-year-olds. The same instinct that lets a teenager offload a maths problem to AI can, unchecked, extend to offloading emotional processing too: instead of sitting with a hard feeling or bringing it to a person, the chatbot becomes the whole conversation.

The tell is not whether your teenager ever talks to a chatbot about how they feel. It is whether the chatbot has become the only place those feelings go - the same judgement-over-avoidance approach that runs through the wider picture in AI education for teenagers in Australia.

Signs worth watching for

Keep the real supports named and available

Whatever your teenager does with a chatbot, the people who can genuinely help have not changed. Your GP is a reasonable first call for anything that feels ongoing rather than a passing bad week. The school counsellor knows the environment your teenager spends most of their day in, and can act quickly if something needs following up at school. Kids Helpline is a free, 24/7 service built specifically for young people, and it is worth your teenager simply knowing it exists, before they ever need it.

None of this requires a crisis to justify using it. A five-minute conversation with a school counsellor about something small is far better placed than a chatbot that cannot actually do anything with what it hears.

How to bring it up without shutting the conversation down

Ask what they use AI for generally, including the parts that are not homework, and listen without reacting sharply to whatever comes back. If you learn they have talked to a chatbot about something difficult, resist the urge to be alarmed out loud. Say plainly that you are glad they found a way to get the thought out, and ask if they would be open to talking it through with you, or with a school counsellor, or your GP. The goal is to be the next step, not to make the chatbot the last one they ever try.

Common mistakes to avoid

• Panicking at the discovery, which teaches a teenager to hide the next hard conversation rather than share it.
• Assuming the chatbot is inherently dangerous, when the real risk is substitution, not the conversation itself.
• Waiting for a crisis before naming Kids Helpline, the GP or the school counsellor as real options.
• Treating this as a technology problem, when it is really a support and connection problem that happens to involve AI.

The recommendation: let a chatbot be one small, low-stakes outlet if your teenager finds it useful, but never the only one. Keep naming the real people - GP, school counsellor, Kids Helpline - so they are familiar and easy to reach before they are urgently needed, and watch for the signs above rather than the fact of AI use itself. A calm, connected teenager who occasionally talks to a chatbot is not a concern. One who has quietly replaced every human conversation with it is, and that is worth acting on early.