You're standing near the back of the room at a Year 10 subject selection evening and someone you know from basketball drop-off catches your eye and says "can I ask you something?"
And then it's the same question I get every time. "What are you telling your kids about all this AI stuff? Because I have no idea what to do."
I've had a version of that conversation with more than two dozen parents in the last few months. All from different schools, different year levels, different backgrounds, same quiet worry underneath all of it. I'm spending a fortune on this. I'm making decisions about subjects and tutors and degrees. And I'm not sure any of it is focused at the right thing anymore.
So this first edition is the answer to that question. The honest version, the one I'd give you if we were standing near the back of the hall together.
Here's the part that's hard to feel until you're inside it.
For most of our working lives, technology has changed things in a way that you could roughly plan around. New tools showed up, you learned them, the job changed a bit, you adapted. The shape of a career, pick a field, pay your dues, build your skills, climb, that basically held.
What's happening now is not that. The thing that's different is the speed of change. The systems I'm using today are wildly more capable than the ones I was using twelve months ago. And the systems I'll be using twelve months from now will make today's look like toys. That's not hype or a sales pitch. That's just what it looks like from the inside.
Most people, very reasonably, are pattern-matching this against every other tech wave they've lived through. The internet. Smartphones. Social. Each of those took ten or fifteen years to really reshape the world. So the instinct is "young people have time, we can wait and see."
I don't think that instinct is right this time. The honest version is that the next three to five years are going to reshape more careers, more quickly, than the last twenty. And the kids who are in Years 9, 10, 11 right now are walking straight into that wall. They don't get to wait and see. We don't get to wait and see on their behalf either.
I'm saying this because if you don't feel that in your stomach, the rest of what I'm about to tell you won't make sense.
OK, so what does that mean in practice. Here's what I'm doing.
Everything. Homework. Research. Hobbies. Side projects. Whatever they're curious about. I want them to develop a reflex, "can AI help me with this?", that fires before anything else. The kids who hesitate to use it are going to find themselves out-paced by ones who don't.
I know that probably cuts against what their school is telling them. A lot of schools are still in defensive mode, still saying "don't use ChatGPT, that's cheating, write it yourself." I understand why. I just don't think it's the right call for what's coming. The world your kid is walking into is one where using AI is the default, and choosing not to is the strange choice.
This is the one I want you to really hear. The trap with AI is that it's so capable it makes the small stuff feel impressive. Your kid asks it to write a story and out comes something that looks great. Tick, done, dopamine hit, move on. That's the trap.
The win is using AI to attempt something that would have been completely out of reach for a teenager three years ago. A real thing. Build an app for a problem at your school. Start a small marketing experiment for a local business. Learn a piece of a field you have no business being in yet. There's a story going around at the moment about a guy who used AI to help work through a cancer vaccine concept for his own dog. That's the level of ambition I want my boys thinking at. I don't expect them to cure cancer. But the attempt is where the skills get built.
I'd rather my kids try something three sizes too big and get 40% of the way there than do the easy version perfectly. The first one teaches them what's possible now. The second one teaches them nothing.
And this is the harder conversation.
Here's what I see happening, and what I think most parents are missing.
AI can give your child the answer to almost any question they bring it. Quickly, confidently, in better English than most adults can write. The first time it happens it feels like magic. The second time it feels normal. By the tenth time it feels like progress.
It isn't.
What's happening is that your kid invested two sentences and a few cents of compute, and got back something that looks like an outcome. Their brain registers a hit of I did a thing. But they didn't do the thing. They asked for the thing. There's a difference, and the difference is the entire game.
The traditional way you got good at something was by going through what I think of as the messy middle. The bit where you don't know what you're doing, you try, you get it wrong, you try again, you slowly build a feel for it. That messy middle is annoying and slow and it's also where every skill you have was built. AI makes it possible to skip the messy middle entirely. And if you skip it, you don't end up with the skill. You end up with the answer, which is a completely different thing.
The output of skipping it is what I've started calling slop. Slop is the assignment that looks fine and reads fine and gets a fine mark but the kid who handed it in could not defend a single sentence of it if you asked them. Slop is the project that looks finished but the kid has no idea how it works, what the tradeoffs were, or what they'd do differently next time. Slop is the dopamine hit of having produced something without any of the understanding that's supposed to come with it.
And slop is what AI produces by default. If your child uses AI without thinking, slop is what they get. Worse, they get the same slop every other student in the class is also getting, because they're all asking the same model the same kinds of questions. The work becomes easier and less differentiated at the same time. Everyone clears the bar. Nobody stands out.
The kids who win from here are the ones who use AI as a power tool but still do the thinking. They know what good looks like because they've developed taste. They can look at what the model gave them and say "that's not quite right, here's what I'd change, here's what's missing." They can defend the work because they understand it. They use AI to be far more ambitious than they could have been alone, but they don't outsource the part of their brain that knows whether the output is any good.
That's the muscle. Taste and judgment, the ability to look at a finished thing and know whether it's done. AI doesn't build that muscle. Used the lazy way, AI lets it atrophy completely.
AI is a superpower that cuts both ways. It lets you become far more ambitious than you ever could have been on your own, but only if you invest the effort. If you don't, it just produces the same slop everyone else is producing, and you end up worse off than if you'd never had it, because you didn't build the underlying skill either.
That's the line I want you to reinforce for your kids.
I'll do this part quickly because I know it's the bit you came for.
I'm not going to tell you law is dead, or accounting is dead, or any of the other clickbait versions of this conversation. I honestly don't know which specific careers will look like what in ten years or even five, and anyone who tells you they do is selling something.
What I am confident about is the shape of what's changing. The technical heavy lifting in a lot of professional work, the parts that used to require years of grinding through detail to get good at, is going to get cheaper and faster very quickly. The thing that becomes more valuable is the layer above that. Knowing what to build. Knowing what the client needs. Having taste. Being able to orchestrate, judge, decide.
So when I talk to my own kids about subjects and direction, I'm not pushing them toward or away from any specific field. I'm pushing them toward developing the meta skills. Can you think clearly. Can you write well. Can you figure out what a problem is before jumping to solve it. Can you tell when something is good and when it isn't. Those skills travel. The specific technical pieces matter much less than they used to, because the model can fill those in. The judgment can't be filled in.
If your kid is set on a particular path, medicine, law, engineering, whatever, I'm not going to tell you to talk them out of it. I'd just make sure they're going in with their eyes open, treating AI as a core part of how they're going to work in that field, and developing the judgment layer alongside the technical one. The kids who do that are going to be fine. The kids who don't are the ones I'd worry about, regardless of which path they pick.
The dog vaccine story is worth a Google. I'm not saying your child is going to cure cancer, but it's the clearest example I've seen recently of an ordinary person using AI to attempt something that would have been completely out of reach a few years ago. That's the level of ambition we should be modelling for our kids.
The "AI detector" arms race at schools is mostly noise. The detectors don't really work, the kids know the detectors don't really work, and the energy spent on the cat-and-mouse would be much better spent teaching them how to use AI well. If your kid's school is leaning hard into detection, that's a signal about where they are on the curve.
Free tools are good. Paid tools are way better. If your kid is serious about a project, the difference between the free version of a major model and the paid version is huge. It's the difference between a clever assistant and a capable collaborator. Worth the twenty to thirty bucks a month if they're using it.
One question to sit with, or to bring to the dinner table this week:
"Show me something you did with AI this week that you couldn't have done without it, and tell me what you'd change about it."
That one question does a lot of work. The first half tells you whether they're being ambitious. The second half tells you whether they're still doing the thinking.
If they can answer both halves well, they're on the right track and you can stop worrying as much as you have been. If they can't answer the first half, push them toward something bigger. If they can't answer the second half, that's where the conversation needs to go.
That's Edition 01. If something in here landed, or if you want to push back on any of it, I want to hear it. The next edition gets sharper because of what I hear back from this one.
The Parent Edit lands roughly every other week. No fluff, no AI hype, just the honest version of the conversation parents keep asking me for.
Get on the listOr forward this to someone who needs it.
What’s this looking like in your house?
Share your experience. Comments are moderated.