We gave our kids unsupervised internet access earlier than many parents would, and it wasn’t entirely by design. When you’re home-educating, the internet stops being optional fairly quickly. Curriculum resources, research, educational videos, typing practice, language learning apps — the screen is just there, and drawing a hard line around it starts to feel increasingly arbitrary when half the lesson plan depends on it.
What we focused on instead was making sure they knew what they were doing when they were on it.
Starting with the basics: online safety
Before independent internet use, they need to understand what the internet actually is — not technically, but socially. What a digital footprint is. What private information means in practice, not just in theory. Why someone online asking personal questions is different from someone in person asking the same questions.
The best structured resource we’ve found for this is the Digital Citizenship course by Common Sense Media. It’s designed to run from primary through to secondary age, building on the same concepts year on year, and it’s set up as ready-to-use lesson plans, which makes it easy to slot into a home-ed curriculum without much adaptation. The ethical questions it raises — about online behaviour, about how you treat people behind a screen, about what you share and with whom — are also useful starting points for writing assignments or wider discussions if you want to extend them.
We used it loosely rather than working through it systematically, dipping in as things came up. That works fine. The point isn’t to complete the course; it’s to have the conversations.
Research: teaching them how to find things out
The other half of internet literacy is knowing how to use the internet to actually learn something, which turns out to be a skill in itself.
The approach that worked best for us was searching alongside them, at least to start with. Not to show them what to search for, but to show them how: how to refine a search term when the first results aren’t useful, how to assess whether a source is worth reading, what to do when two sources say different things. These are habits that take time to build and are much easier to build out loud, together, than to explain in the abstract. That’s the conversational curriculum in action, if you want a framework for it.
One bonus that came out of this was that we started keeping a list of questions that came up during the week — the random ones, the “why does” and “what would happen if” kind — and using them as the basis for a search session. It generates genuinely interesting material, usually throws up something worth bookmarking for a future lesson, and it teaches the process without it feeling like a lesson about the process.
The order we approve for research is books first, then the wider internet, then AI. Not because the internet or AI are less reliable exactly (though they can be), but because the process of looking something up, hitting a dead end, trying a different angle, finding something unexpected along the way, is part of how you actually learn a subject. Skipping straight to a tidy AI summary misses all of that. There’s a time for the summary and it’s not usually the beginning.
Setting the ground rules
Rather than device controls, which tend to create frustration without building the underlying understanding, we’ve found that explicit house rules work better for us. Which sites they can use freely, which need checking with us first, what the rules are around signing up for things, sharing their name, and turning on chat functions. Having that conversation once, clearly, and revisiting it as they get older is more useful than a filter that they’ll eventually work around anyway.
Screen time is harder to manage when the curriculum runs partly online. What we’ve tried to make habitual is balance: one online activity, one offline one. Not rigidly, but as a default direction.
The AI question
AI sits alongside internet use, but it’s different enough to need its own conversation. The same principles apply — what it’s good for, what it isn’t, and why the distinction matters — but the specifics are different enough to need their own post. Here’s how we think about AI in our homeschool.
