I think that’s the right place to start, and not because your kids don’t matter. It’s because you can’t model good judgment about a tool you’ve never really used. The fastest way to become a useful guide isn’t to study the dangers from the outside — it’s to build a little fluency of your own, on low-stakes stuff, until you can feel where the technology is strong and where it quietly lets you down.

So here are four ways I actually use AI in the unglamorous business of running a family. Each one comes with the prompt pattern I’d hand you, what it’s genuinely good for, and the place it’ll burn you if you trust it blindly. That last part matters most. The single habit that separates people who use AI well from people who get used by it is simple: never let it be the last word on anything that matters. Treat it as a sharp, fast, occasionally confident-and-wrong assistant — never as an authority.

1. Decoding the thing you don’t have time to understand

Your kid comes home with a diagnosis, an unfamiliar term from a teacher, an IEP acronym, a form full of jargon. You’ve got twenty minutes and a search engine full of garbage.

This is where AI earns its keep — not by telling you what to do, but by translating. Try:

“Explain [term] to me like I’m a smart adult who isn’t a specialist. What does it actually mean, what questions should I be asking the doctor/teacher about it, and what would you want to rule out or clarify? Don’t give me advice — give me better questions.”

What it’s good for: turning a wall of jargon into plain language and a list of sharp questions to bring to the actual professional.

What it is not for: deciding anything medical or clinical. It does not know your child, it cannot examine anyone, and it will state wrong things with total confidence. The move is to walk into the appointment understanding the vocabulary and armed with good questions — not to replace the appointment. Verify everything important against a real clinician.

2. Drafting the email you’re too angry to write

Some of the most useful AI moments in my house have nothing to do with my kids using it. They’re me, at night, needing to send an email to a teacher or an administrator about something that’s got me genuinely heated — and knowing that the version I’d write in that state will make things worse.

“Here’s a situation and the rough email I want to send. [paste your angry draft.] Keep my point and my firmness completely intact, but rewrite it so it’s calm, specific, and hard to dismiss. Tell me which lines would read as hostile and why.”

What it’s good for: taking the heat out of the language while keeping the spine. It’s a translator between what you feel and what lands well. The “tell me which lines read as hostile” part is the real value — it surfaces the stuff you can’t see when you’re activated.

What to watch: don’t let it sand off your actual position to be polite. Read the result and put your edges back where they belong. The goal is you, on a good day — not a beige committee memo.

3. Prepping for the hard conversation

The talk about the grades. The friend group. The thing you saw on their phone. I used to walk into those cold and improvise, which mostly meant I led with the lecture and lost them in the first thirty seconds.

“I need to talk to my teenager about [topic]. Help me think it through before I do. What’s likely going on from their side that I might be missing? What are three ways this conversation could go, and how do I keep it from turning into a lecture? Ask me questions if you need more context.”

What it’s good for: rehearsal. It plays out the conversation, points at the version of events you’re not considering, and — if you let it — calls you on the move where you stop listening and start delivering a verdict.

What to watch: it doesn’t know your kid. The value is the rehearsal, not the script. Walk in with a better-prepared head, not a memorized monologue.

4. Thinking a real decision out loud

School options. Whether to allow a phone. Screen-time and AI-time policies that won’t fall apart in a week. These are the decisions where I most want to think clearly and most often don’t, because they’re tangled up with worry.

“I’m weighing [decision]. Here’s my situation and the factors I’m juggling. [lay it out.] Don’t tell me what to do. Lay out the real tradeoffs, name the things I might be underweighting because I’m anxious about them, and ask me the questions a thoughtful friend would ask before I decide.”

What it’s good for: structure. It turns a knot of worry into a clear list of tradeoffs and forces the quiet assumptions into the open. The “don’t tell me what to do” instruction is doing a lot of work — left to its own defaults it’ll hand you a confident recommendation, and a confident recommendation is exactly what you don’t want when the right answer depends on your specific family.

What to watch: it has no stake in your kid’s life. It’s a thinking partner, not a decider. You decide.

The one habit underneath all four

You’ll notice the same move in every prompt above: I ask it to give me better questions, clearer tradeoffs, and a second perspective — and I never ask it to be the authority. That’s the whole game. AI is genuinely excellent at the middle of the work — translating, drafting, structuring, rehearsing — and genuinely unreliable as the final judge of anything that matters. Keep it in the middle, keep yourself at the end, and it becomes one of the most useful tools you’ve got for the actual labor of parenting.

Once that habit is yours, something nice happens: your kids watch you use AI with judgment instead of fear, and that — not any rule you post on the fridge — is the thing that actually teaches them.

That’s the stuff I want to dig into here, one piece a week. If you’ve got a workflow of your own that’s earned its place, hit reply and tell me — some of the best ones in my house started as someone else’s tip.

(For anyone who wants to go further, a few of us keep a running, copy-paste library of these prompts in a small parents’ community. Entirely optional — the newsletter stands on its own.)

— Peter

One week it’s a headline telling me AI is quietly dismantling my kids’ ability to think. The next it’s someone promising that an AI tutor will fix everything school got wrong. Both can’t be right, and honestly, neither matches what I actually see at my own kitchen table.

So I’m writing the thing I kept looking for and couldn’t find.

Here’s where I’m coming from. I’ve spent 20-plus years in tech, from Intel to AMD, building the security that sits underneath the systems all of this AI now runs on — the deep, unglamorous infrastructure meant to keep it from failing in ways you’d never see coming. I’m not telling you that to flash a badge. I’m telling you because it’s why I’m hard to sell to and slow to panic: I spend my days looking at how these systems actually break, instead of how the demo says they work.

And I’m a parent. One of my kids used these tools to start and run a small service business before he was old enough to drive. I watched that happen up close — what the AI did, what he did, where it helped, and the exact moment it would have led him off a cliff if no one were paying attention.

That combination — someone who knows what’s under the hood and is living the parenting version of this in real time — turns out to be rare. Most of the loud voices have one half or the other. The technologists don’t have a teenager negotiating with a chatbot at 11pm. The parenting writers haven’t seen how the machine actually works. I keep ending up in the gap between them, and that gap is where I want to write from.

What this newsletter is actually about

Here’s the part most parenting-and-AI content gets backwards. It treats you, the parent, as a gatekeeper — someone whose job is to set limits on a thing you’re standing outside of. I don’t think that works. You can’t usefully guide your kid through something you’ve never genuinely used yourself. The parents I trust most on this aren’t the ones with the strictest rules. They’re the ones who’ve built real fluency of their own, and parent from there.

So this covers three things, in this order:

1. How you can actually use AI in your own life. Before anything about your kids — the boring, high-leverage stuff. Drafting the hard email to the teacher when you’re too angry to write it well. Getting your head around a medical question at midnight without three hours of bad search results. Thinking through a real decision out loud with something that pushes back. If you build this muscle for yourself, everything downstream gets easier.

2. How a family can use AI together. Side by side, out loud, with the kids watching how you handle it — when you trust it, when you double-check it, when you walk away. That modeling does more than any rule you’ll ever post on the fridge.

3. How to raise kids who are capable with it, not dependent on it. The part everyone else leads with — but grounded in how the technology really behaves, not in vibes or fear.

What you’ll get, and what you won’t

One essay a week, mid-week. Long enough to actually say something — eight or ten minutes to read — and short enough to finish. Now and then a quick note in between when I notice something worth flagging. Every so often, an honest write-up of what my own kid is building and what it’s teaching both of us.

What you won’t get: listicles, breathless tool roundups, affiliate links dressed up as advice, or anything written by a machine. I write all of this myself. That’s sort of the whole point.

And when I don’t know something — which happens — I’ll say so plainly instead of performing certainty. The defaults in this space are hype and dread. I’m trying to hold a third position: take the technology seriously, take parenting seriously, and treat you as an adult who can form your own view.

The honest version of the deal

I’m not going to pretend I have this fully solved. Nobody does — the ground is moving too fast, and anyone who tells you they’ve got a tidy framework for all of it is selling something. What I do have is a particular vantage point, a kid who’s been my unintentional test case, and a strong allergy to both the panic and the hype.

If that sounds like the conversation you’ve been wanting to have, subscribe below and the next one will land in your inbox. There’s also a small community of parents working through this stuff in real time, for anyone who wants to go further than a once-a-week essay — but that’s entirely optional, and the newsletter stands on its own.

Either way: I’m glad you’re here. Let’s figure this out like the grown-ups in the room.

— Peter

Subscribe now