The "internet safety talk" is one of those parenting landmarks that feels important but also, if we're honest, a bit vague. What exactly are you supposed to say? Where do you start? How much is too much for a seven-year-old?
The research on this is consistent: a single conversation doesn't work. What works is a series of ongoing, age-appropriate conversations that grow as your child grows. Children who have talked with their parents about the internet are significantly more likely to come to them when something goes wrong, and that communication pipeline is the most valuable online safety tool you have.
This is your age-banded guide. Save it and come back to it as your child moves through each stage.
How to use this guide: each age band builds on the last. You don't need to cover everything in one sitting. The goal is an ongoing conversation, not a lecture your child sits through once and files away. Little and often, curious and open, not frightening or preachy.
Ages 5-7, the foundation: simple rules, big trust
At this age, children are just beginning to use devices, often for YouTube, games, or video calls with family. They're concrete thinkers. Keep everything simple, positive, and about what they can do as much as what they can't. The goal is to establish that talking to you about the internet is normal and safe.
1. What we keep private online
Your name, address, school, phone number, and photos stay just for us and people we know in real life. Not because the internet is scary, because private things are precious.
Try: "If a game asks for your real name, what do you think we should do?"
2. The ask-first rule
Before clicking on anything new, downloading anything, or joining a new game or app, always ask a grown-up first. Not because you're in trouble, but because some things are set up to trick even adults.
Try: "If you see a button that says 'click here for a free prize', what would you do?"
3. Coming to you when something feels wrong
The most important thing: if anything online makes them feel worried, confused, scared, or uncomfortable, they should tell you immediately, and they will not get in trouble. Establish this now, clearly, repeatedly.
Try: "If something you see online ever makes you feel a bit weird or scared, you can always, always tell me. Even if you think it was your fault. Okay?"
Ages 8-10, the fundamentals: strangers, kindness, and the footprint
At this age, children are often moving into more social online environments, Roblox, Minecraft, group chats, early YouTube content creation. They're beginning to understand consequence. Now you can go a level deeper: why these rules exist, not just what they are.
4. Online people aren't always who they say they are
People online can present themselves as anyone. A 12-year-old in a game might actually be an adult. This isn't about fear, it's about understanding that online identity is easy to invent. The rule isn't "don't talk to anyone", it's "don't share personal information or meet someone you only know online."
Try: "Have you ever met someone in a game who seemed friendly? What do you actually know about them?"
5. Being kind online, and what to do when others aren't
Unkindness online spreads faster than it does in person, because there's no face to see. Being kind online is a real choice. But they'll also encounter unkindness, from strangers, and sometimes from people they know. What do they do? They don't retaliate. They screenshot and tell an adult.
Try: "If someone said something unkind to you in a game, what do you think you'd do?"
6. Your digital footprint, what the internet remembers
Anything they post, comment, or share online can potentially be seen by others and doesn't fully disappear. Not to frighten them, but to help them think before they post. The question to teach them: "Would I be okay if this showed up at school tomorrow?"
Try: "Do you know what a digital footprint is? Let me show you how to find what comes up when you search your name."
Ages 11-12, social media: algorithms, comparison, and identity
At this age, children are either already on social media or asking to be. They're experiencing more complex social dynamics, group chats, exclusion, online friendships that feel very real. The conversations here need to go deeper: how the internet is designed to work on them, and who decides what they see.
7. How algorithms actually work
Platforms show content designed to keep your child watching, not content that's necessarily good for them, true, or kind. This is a design decision, made by engineers. Understanding this shifts the relationship from passive consumer to someone with awareness.
Try: "Have you ever noticed that after you watch one video, you suddenly end up watching loads of similar ones? Why do you think that happens?"
8. The comparison trap, what social media actually shows you
What people post online is a curated highlight reel, not real life. This isn't cynicism, it's media literacy. When we compare our insides (how we actually feel) to other people's outsides (what they choose to show), we always lose. Help them notice when they're doing this.
Try: "After you spend time on TikTok, how do you usually feel? Better or worse about yourself? Have you ever thought about why?"
9. What to do when something goes seriously wrong
Before your child has social media, make sure they know the practical steps for the situations that do happen: seeing self-harm content, being asked to share intimate images, receiving threats, being cyberbullied, or encountering someone suspicious. They need a clear action plan, not just a vague "tell me."
Try: "If someone online ever sent you something you really didn't want to see, or asked you for something that felt wrong, what would you do first?"
Ages 13 and over, autonomy: who you are online, and what you create
Teenagers don't want rules from parents, they want autonomy. The best conversations at this stage treat them as capable of making good decisions when they have good information. Your role shifts from instructor to thinking partner.
10. Creating vs consuming, what you actually want from the internet
The most powerful question for a teenager: how much of your online time is consuming, and how much is creating? The research on digital wellbeing consistently shows that passive scrolling is far more likely to affect mood negatively than active creation. Help them think about what kind of internet user they want to be.
Try: "If you looked at how you spent your last hour online, was it mostly watching and scrolling, or were you actually making or doing something? Which do you prefer?"
One thing that works better than all of these
More important than any single conversation is the environment you create around these conversations. Research published in the Journal of Adolescence found that teenagers are significantly more likely to disclose online problems to parents who respond with curiosity and support rather than immediate worry and restriction.
In practice, this means: when your child tells you something has gone wrong online, your first response matters more than the conversation you planned. Stay calm. Thank them for telling you. Deal with the problem together. That response is what determines whether they come to you next time.
The rule that matters most: "You can always tell me. Even if you think it was your fault. Even if you think I'll be angry. You will not get in trouble for coming to me." Say this. Repeat it. Mean it.
Shield, Protection Mode helps children develop an instinct for what to share and what to keep private, one of the foundational skills behind every conversation in this guide. Conversations land better when children already have the language to think about them.
Emi, Emotion Mode builds emotional vocabulary and the ability to recognise how online environments are affecting mood. That self-awareness, noticing when TikTok is making me feel worse, is exactly what makes the comparison trap conversations land.
Link, Connection Mode builds social and relational intelligence, the ability to recognise the difference between genuine connection and online performance. As children move into more social online environments, this is the skill that protects the quality of their real relationships.
Codey, Logic Mode builds critical thinking, the ability to question, verify, and understand how digital systems work. The algorithm conversation is much easier when a child has already started thinking like Codey: why am I seeing this? Who wants me to see it?
Questions parents ask
When should I start talking to my child about internet safety?
The earlier, the better, ideally before your child begins using any device independently. Research consistently shows that conversations started early and revisited regularly are far more effective than a single "the internet is dangerous" talk. Even children aged 5-6 can understand simple concepts like "we don't share our address online" or "ask an adult before clicking."
What should I talk to my 7-year-old about online?
At 5-7, the most important conversations are about what not to share online (name, address, school, photos), what to do if something makes you feel scared or uncomfortable, the difference between real friends and online strangers, and always asking a trusted adult before downloading or clicking on something new.
How do I talk to a teenager about internet safety without them switching off?
Teenagers respond better to conversations framed around their autonomy and intelligence, not rules and fear. Ask curious questions rather than delivering lectures. "What do you actually enjoy about TikTok?" is a better starting point than "TikTok is dangerous." Validate what they do well online before discussing risks.
What are the most important online safety conversations to have?
Across all ages, the highest-impact conversations are: what not to share online; how to recognise when something online doesn't feel right; what to do when something goes wrong (and knowing they won't get in trouble for telling you); how algorithms work; and how to make active choices about what they consume and create online, rather than just scrolling.
Should I monitor my child's online activity?
Parental controls and monitoring can form a useful layer of protection, but research consistently shows they work best alongside, not instead of, open conversations. Children who feel their online life is entirely surveilled are less likely to come to parents when something goes wrong. The goal is a child who self-regulates because they've genuinely understood why it matters.
For more on the platforms themselves, see our platform-by-platform guide to social media risks. havyn is a children's digital literacy app that gives children aged 5-13 the skills that make these conversations land, and stick.