Can kids safely use AI?

It can feel like you’re constantly playing catch-up as a parent, especially where tech is concerned. Take AI for example. It's quickly become part of everyday life and within easy reach for kids. But what’s helpful about AI? What’s unsafe about it? And how do you equip kids to be smart about using it?

First things first, while tech continues to evolve at a rapid rate, what matters most for parents and caregivers is staying connected. A parent’s biggest influence is still found in their values, boundaries, relationships and regular conversations. So, while it helps to know what’s going on in the world of AI and how kids are using it, it’s even more important to help them navigate it with balance and safety in mind.

AI drives a lot of the tech that we use day to day: suggesting new music we might like, powering voice assistants and other smart devices, predicting traffic and travel time, optimising online searches and so much more.

Then there’s generative AI – tools that are in the user’s control and can create new original content. Generative AI can summarise information quickly, explain concepts in different ways and prompt new ways of thinking. It can also support creativity – from giving kids the confidence to ‘get started’, to generating new ideas for design, music or even story characters.

These can be real positives – but keep in mind, kids will still need clear boundaries and strong adult support.

Like all power tools, AI comes with risks, especially for young users who are still developing critical thinking skills.

AI CAN GET IT WRONG

It’s easy to over‑trust AI because the responses feel authoritative and fast, and kids simply don’t have the worldly experience to determine fact from fiction.

Parents – encourage kids to use AI for ideas and prompts rather than finished answers, and show them how to ‘sense-check’ information. Ask questions like: “Is this suggestion fair, or could it be missing something important?”, and “What would we want to double-check before we use this information?”

AI CAN HIJACK LEARNING AND CREATIVITY

Even for adults, it's tempting to let AI do too much of the thinking and shortcut learning and originality. When kids are still learning how to learn, this overreliance on AI can be particularly problematic, especially if it crosses the line into cheating or plagiarism.

Parents – remind kids that AI is a computer tool designed to look like it can think. It's great for prompting thought and breaking down complex topics, but not for producing work word for word. It can't replace the depth, creativity and judgement of the human brain so let’s not over-value or over-rely on what it can offer.

AI DOESN'T TRULY CARE

AI companions and chatbots can be handy tools, but they should never become a young person’s closest friend – or the ‘agony aunt’ they go to first. Psychologists are quick to warn that AI companions are particularly problematic because they are designed to be endlessly affirming, and miss the healthy friction found in human relationships that teach empathy, conflict resolution and responsibility. The advice they give can also be flawed or misguided.

Concerningly, AI companion conversations can sometimes lead to explicit sexual, abusive or self‑harm themes. More information to help you understand the risks of AI social companions is available at Netsafe.

Parents – help your kids to understand the limitations of chatbots; that they can sound human, but they’re not really friends or experts.

AI CAN RISK DATA PRIVACY

Sharing any personal information online creates a digital footprint that's hard to control or erase. Many AI tools store conversations, using them to train models or improve services, and data handed over can be linked, analysed and used in the future. Kids often assume chats are private and temporary when in fact they’re not.

Parents – remind your kids to treat AI like a stranger online. It’s easier to withhold personal information than to try to rein it back in if you’ve overshared online!

AI CAN CREATE HARMFUL AND ILLEGAL CONTENT

Some AI apps make it easy for young people to encounter, or even generate, content that crosses a legal line. It can expose kids to realistic sexual and violent images, videos and deepfakes that look like real people – and it can be used for bullying, humiliation, sextortion and blackmail. This type of dodgy content is incredibly distressing for a young person to encounter and should be taken very seriously.

In Aotearoa, some sexual or violent content (especially involving young people) is illegal to create, share or even keep, even if it’s been generated by AI and no 'real person' posed for it. More information on the risks and harms of objectionable AI content, and where to seek support and report it can be found at New Zealand Classification Office Te Mana Whakaatu.

If you haven’t already – add ‘AI’ to the list of ongoing conversations you’re having as a whānau around device use, online life and digital wellbeing. Agree on the types of AI tools that you’re comfortable with the kids using, bearing in mind their age and development, and considerations like safety and privacy.

Ensure the boundaries are clear and agreed on. For younger kids, this could look like only using chatbots alongside a parent or caregiver (co-viewing and co-playing). For older kids and teens, it could be agreeing that you may check their chatbot activity from time to time, put limitations on access and initiate regular check‑ins if tools like Claude, ChatGPT or Midjourney are being used for schoolwork or study.

Our downloadable First Phone Agreement is a helpful tool for getting agreement on rules and boundaries upfront.

You know your kid best! If they are feeling confident that AI is supporting their learning or creativity, they’ll be able to explain how they used it and what they’ve created.

Alternatively, if they’re hiding their screen or becoming over reliant on AI tools or companions, it may be time to step in. Start with curiosity, not sneaking around or looking for gotcha moments. Kids open up when they feel safe – and how you respond makes all the difference.

If something does go wrong (i.e. they run into harmful AI content, they are subject to sextortion or they get pulled into distressing chatbot interactions), stay calm and avoid putting the blame on your child. Save any evidence and reach out to your child’s school, Netsafe or the NZ Police for further help.

The goal is to make AI use something kids can talk about openly and confidently. Clear boundaries help, but so does reassurance that curiosity is normal, questions are welcome and you’re there to figure this out together.

For more on keeping your kids safe with AI, check out: * Top tips for parents and carers - Safer Internet UK

• Children and chatbots: Your questions answered

• What is Grok?

• AI social companions and youth: What parents need to know