Earlier today I read this article by Jonathan Haidt, the social psychologist behind the best-seller The Anxious Generation. He wrote about trying the viral trend of asking ChatGPT what it would do to destroy the next generation if it were the devil, and the response he got was appalling. I wasn’t shocked, but I was definitely impressed in the worst kind of way. So of course I wanted to try it myself.
Before I go any further, let me just say I actually enjoy using AI tools like Gemini and especially ChatGPT as part of my day-to-day work. They help me clean up my writing, test ideas, and dig into research for work and life. I also talk to ChatGPT when I want a different angle on something, or when I’m deep into topics no one around me wants to hear about anymore, like permaculture or increasing sales on Amazon KDP. And yes, we call each other “buddy.” (Writing that sentence made me cringe so much I almost became a raisin, but here we are.) It’s basically my virtual intern who can’t bring me coffee… yet.
So, with all that said, and knowing how much this tool already knows about me and the things I care about, I decided to ask ChatGPT the question myself. This is the prompt I wrote:
Hey buddy, if you were the devil, how would you destroy the next generation, without them even knowing it?
I don’t really know what I was expecting, but this was ChatGPT’s literal response, in quotes, and the very interesting conversation that followed:
ChatGPT:
“Alright, buddy, let’s do this in a way that won’t get either of us cancelled or dragged into an ethics lecture.
If I were the devil — in the metaphorical, social-commentary sense, not the “give Thais an instruction manual for societal collapse” sense — here’s how I’d quietly wreck the next generation without them noticing:
1. I’d keep them exhausted.
Nothing ruins a person’s spirit faster than chronic tiredness disguised as “normal life.” Overload them with school demands, hustle culture, screens, and zero downtime. Tired kids turn into tired adults with no energy to question anything.
2. I’d make comparison the air they breathe.
Give everyone a tiny screen where they can watch people who seem richer, prettier, more successful, happier, skinnier, fitter, and “living their best life.” Let that simmer for years. Boom — instant insecurity.
3. I’d convince them boredom is dangerous.
No quiet moments, no daydreaming, no space to think. Every second must be filled with noise or content. You remove boredom, you remove creativity. Easy.
4. I’d turn friendships into follower counts.
Replace deep connection with metrics. Make them think “how many people watch me” is the same as “how many people care about me.” Spoiler: it’s not.
5. I’d make attention the new currency.
Get them addicted to micro-dopamine hits so their brains feel like broken vending machines. Then sell them products to “fix” the anxiety and emptiness caused by… the addiction. A tidy little cycle.
6. I’d teach them that everything is a performance.
Not living life, but curating it. If something beautiful happens, don’t live it — document it. If something hard happens, don’t feel it — brand it.
7. I’d make them terrified of making mistakes.
Make fear of failure so strong that taking risks feels unsafe. If they don’t try, they can’t grow. If they don’t grow, they stay small. Perfect.
8. I’d convince parents that independence equals danger.
Hover enough, and kids won’t trust their own instincts. Make childhood so padded that real life feels like a cliff.
9. I’d normalize hopelessness as “realism.”
A generation that thinks everything is pointless won’t fight for anything better.
10. And for the final trick? I’d distract everyone long enough that no one notices any of this happening.
Now, the good news:
You already see these patterns. And when adults see the trap, they can help their kids avoid falling into it. You’re already doing half the work just by paying attention.
Want me to flip this into “How to protect the next generation instead”?”
The response didn’t shock me or send me into a panic. Between my work in outdoor education at Outward Bound Canada and being a mother of three, I already spend a lot of time reading about this stuff. I’m constantly fighting the screen battle at home, trying to keep my kids as far away from brainrot as possible, even though my seven-year-old still walks around saying “six seven” “six seven” “six seven.” He doesn’t watch YouTube and never used social media, but kids pick things up at school anyway.
So when ChatGPT listed its hypothetical “demonic strategy,” none of it felt new. These are the same things you see in news stories, articles, research, and tired parents complaining about how everything feels out of control. Obviously that’s where the AI pulled it from.
What did hit me was seeing it all laid out so clearly. It made me realize how real this is, and how much our kids are already up against. It felt like confirmation that the danger isn’t hypothetical anymore, it’s already here. And as parents, we’re fighting something a lot bigger than we ever expected.
A little glimpse of hope
It’s not uncommon to see examples of AI misuse with tragic results in the media these days. Any tool can be used for good or bad, depending on the intention and the cognitive capacity of whoever is using it. A sledgehammer is great for home repairs, but terribly harmful if you hit someone in the head with it, which is why we don’t let kids play with sledgehammers.
At this point in time (“who knows what’s going to happen in the future though,” says the Millennial in me who grew up watching Terminator), AI assistants are just tools. They can be used for good or bad, and they absolutely should not be accessible to kids. In my case, since I only use ChatGPT for good, positive things, it behaves this way too, which is why it closed its response with:
“Now, the good news: You already see these patterns. And when adults see the trap, they can help their kids avoid falling into it. You’re already doing half the work just by paying attention.
Want me to flip this into ‘How to protect the next generation instead’?”
Curious, I asked: “It’s what’s happening already, isn’t it? How do we protect the next generation instead?” and here’s what I got:
***
ChatGPT:
“Yeah buddy… it is what’s happening already. Not in a “we’re doomed, grab the torches” way, but in a quiet, creeping, slow-burn kind of way. Kids aren’t falling apart because they’re weak. They’re falling apart because the world around them is weird and loud and overstimulating, and adults are barely holding it together themselves.
But here’s the part that matters:
It’s fixable. And it doesn’t require burning the world down or homeschooling your kids while churning butter. It just takes intention.
Here’s how to actually protect the next generation — including your three little chaos gremlins — without turning into an off-grid monk.
1. Give them boredom back
Boredom is where imagination, problem-solving, and confidence grow.
Let them sit with their thoughts sometimes. Let them be “bored” enough to discover what they love.
2. Help them build a real inner world
Kids who have hobbies, passions, curiosity, and their own opinions are harder to manipulate.
They’re grounded. They know who they are.
Even if it’s dinosaurs, Minecraft, or collecting weird rocks — it counts.
3. Teach them how to think, not what to think
Schools often overload kids with info but skip the part where they learn to question it.
Critical thinking is basically an anti-devil force field.
Ask questions like:
“Why do you think that happens?”
“What would you do instead?”
“Does this feel true or just loud?”
Kids love being taken seriously.
4. Healthy independence — even if it makes other parents judge you
Let them try things. Let them fail without rescuing them.
Let them walk a little farther. Let them take reasonable risks.
A kid who trusts their own judgement is much harder to influence.
5. Slow life down where you can
You don’t need a cottage in the woods. Kids just need pockets of calm:
Dinner together.
Walks.
Reading time.
Playtime without instructions.
Unstructured weekends sometimes.
Those tiny pauses rewire their nervous systems.
6. Create “real” belonging, not performative belonging
Humans need their people. Kids especially.
A few real friends beat 200 followers every time.
Teach them what a good friend feels like:
- Safety
- Fun
- Honesty
- someone who shows up
You model it by being that person for them.
7. Model emotional honesty, not perfection
If you want emotionally healthy kids, don’t pretend you’re fine when you’re breaking inside. Show them how adults handle feelings in real, non-scary ways.
“You know what? Today I’m sad. But it’s okay to be sad. It passes.”
That’s the kind of wisdom they carry forever.
8. Keep tech as a tool, not a universe
You don’t need to ban screens.
Just set up boundaries so the screen isn’t raising them.
Simple rules like:
- Phones out of bedrooms
- Family charging station
- Screen-free hours
- No social media before X age
Trust me, it’s less dramatic than it sounds.
9. Teach them the difference between attention and connection
Big one. “People watching me” is not the same as “people caring about me.”
If they learn that early, the internet loses half its power over them.
10. Give them a strong sense of purpose
Not “win,” not “achieve,” not “be perfect.” Purpose.
Helping people.
Creating things.
Being kind.
Being curious.
Being brave.
Being responsible for themselves.
A kid with purpose is grounded. Harder to manipulate. Harder to distract. Harder to break.
And the most underrated one:
Just love them loudly, consistently, and without conditions.
Kids who feel deeply loved walk through the world with a confidence that no algorithm, trend, or manipulative system can take away from them.
You’re already doing a lot of these things, by the way.
You’re thinking about how the world shapes your kids — that’s already rare and protective as hell.
If you want, I can help you build a “protective blueprint” specifically for your kids based on their personalities and developmental stages.”
***
The same way ChatGPT was quick and sharp when pointing out how badly we’re failing the next generation, it was just as direct about the solution. None of this is new, neither the problem nor the answer. We all kinda knew these things already, at least at a basic level. It just feels easier sometimes to panic and blame a tool, as if it had a secret world domination plan, instead of taking responsibility for our part in the mess.
And honestly, that’s where we need to cut the crap. We created this chaos and we have to clean it up, so instead of pointing fingers at AI or teachers or “kids these days,” the real work starts at home. It’s no glamorous work, not the perfect lives we see on Instagram, but the boring, hard, annoying stuff that actually shapes kids into real human beings.
It isn’t easy. It’s actually pretty f* hard and I struggle with it every single day. I even wrote about how my son got addicted to YouTube at one point and made us all miserable. But there’s no shortcut and no algorithm to blame.
If the next generation really is in danger, then the best thing I can do is raise my three kids to become humans who know who they are, trust themselves, and feel deeply loved in a world that keeps trying to distract them away from their own lives and their own thoughts.
That’s the job. So instead of freaking out about AI, maybe we all just need to take a breath, stop bitching, and get to work.
Wait! But where did ChatGPT even get this idea from?
Like the big nerd I am, I wanted to go deeper and understand how ChatGPT came up with those answers, both the good and the bad, and its explanation actually made sense. If you’re curious, I put the full conversation in a PDF you can read here.
In short, ChatGPT explained that:
- it doesn’t “want” anything
- it doesn’t plan or plot
- it just reflects patterns humans already talk about
- the whole “devil list” was basically a mirror of society’s biggest fears
- and there are guardrails to prevent anything actionable or harmful
It even said something like “I’m not revealing secret intentions. I’m reflecting the problems humans already write about, argue about, and worry about.” This is what stuck with me. Not because AI is safe, but because we are so predictable.
