If you walk into a high school hallway today, you’ll notice something eerie. It isn’t the noise that hits you—it’s the silence. Hundreds of kids are sitting against lockers, necks craned at a 45-degree angle, eyes locked onto glowing rectangles. They’re together, but they are completely alone.
Social psychologist Jonathan Haidt has spent the last few years ringing every alarm bell he can find about this "Great Rewiring of Childhood." He’s famous for his book The Anxious Generation, where he argues that we essentially traded a play-based childhood for a phone-based one, and the results have been disastrous for mental health. But now, there’s a new predator in the room.
Artificial Intelligence.
Honestly, we aren't ready. Haidt’s latest warnings suggest that if social media was a tremor, AI is the massive earthquake that could permanently shift how students learn—or stop learning altogether. It isn't just about cheating on an essay anymore. It's about the fundamental erosion of the human mind's ability to do hard things.
Jonathan Haidt on the Impact of AI in Student Learning: The Attention Trap
Haidt’s core thesis is pretty simple: kids are being "under-protected" in the virtual world while being "over-protected" in the real one. We don't let them walk to the park alone, but we let them talk to strangers and algorithm-driven bots for six hours a day.
When it comes to AI, the danger isn't that the tech is "evil." It’s that it is too convenient.
Learning requires friction. You’ve probably felt this yourself. When you struggle to solve a math problem or spend an hour trying to structure a paragraph, your brain is actually growing. Haidt points out that AI removes that struggle. Why spend three hours researching the Causes of the French Revolution when a chatbot can spit out a perfectly structured summary in four seconds?
📖 Related: What Was Invented By Benjamin Franklin: The Truth About His Weirdest Gadgets
The problem? You didn't learn anything. You just managed a process.
The Death of "Productive Struggle"
In his recent talks and writings, Haidt has been vocal about how AI could "take all the pathways of harm from social media and multiply them." In a 2025 interview, he noted that 40% of American two-year-olds already have their own iPads. By the time these kids get to middle school, they aren't just using AI to check their work; they're using it to bypass the thinking process entirely.
He calls this a "potentially unsolvable problem for education."
Teachers are basically in a digital arms race. On one side, you have students using increasingly sophisticated LLMs to generate assignments. On the other, you have schools buying "AI detectors" that barely work and often flag innocent students. It's a mess. But for Haidt, the academic integrity part is actually the least of our worries. The bigger issue is cognitive atrophy.
If a student never has to sit inside "uncertainty," they never develop judgment. AI provides instant, confident answers. But real life doesn't come with a prompt box. If we raise a generation that reaches for a bot every time they hit a wall, we are raising a generation that can't lead, can't innovate, and honestly, can't think for themselves.
Why "Phone-Free" is Only the First Step
You’ve probably heard Haidt’s "Four Norms" by now. They’ve become a bit of a manifesto for parents and educators:
👉 See also: When were iPhones invented and why the answer is actually complicated
- No smartphones before high school.
- No social media before age 16.
- Phone-free schools (the "bell-to-bell" rule).
- Way more unsupervised play.
But how does this map to the AI explosion?
Haidt argues that we have to get the "delivery device" out of the classroom first. If a kid has a smartphone in their pocket, they have a 24/7 portal to an AI that will do their thinking for them. By pushing for phone-free schools—meaning phones are in lockers or magnetic Yondr pouches—schools can at least create a "sanctuary of attention."
It's about rebuilding the "human" part of school.
The Gendered Divide in Digital Harm
One of Haidt’s more controversial—but data-backed—points is that tech hits boys and girls differently.
For girls, social media (and now AI-generated social pressure) leads to "social deprivation" and a comparison trap. They are looking at AI-filtered versions of their friends and feeling like they don't measure up. For boys, the risk is different. They tend to disappear into "virtual worlds"—gaming, pornography, and now, AI companions.
Think about that for a second. We are seeing a rise in "AI girlfriends" and "AI friends" that are designed to be perfectly agreeable. Haidt warns that this stunts social development. If a boy’s only "friend" is a bot that never disagrees with him, how is he supposed to handle a real-world conflict with a classmate or a future boss? He’s going to be "socially fragile."
✨ Don't miss: Why Everyone Is Talking About the Gun Switch 3D Print and Why It Matters Now
The Counter-Argument: Is Haidt Just a Luddite?
Look, not everyone agrees with him. Critics like Candice Odgers, a psychology professor at UC Irvine, argue that the link between tech and mental health is "correlative," not "causal." They say we’re blaming the tools for a much broader societal failure.
Educators also point out that AI isn't going away. If we don't teach kids how to use it, aren't we just making them "digitally illiterate"?
Haidt’s response is usually some version of: "Sure, but let them grow up first."
He isn't saying we should ban AI forever. He’s saying we shouldn't give a 12-year-old a tool that mimics human consciousness before their prefrontal cortex has even finished developing. We don't let kids drive cars at 10 just because "cars are the future." We wait until they have the maturity to handle the power.
Actionable Steps for Parents and Teachers
If you're feeling overwhelmed by the "AI-pocalypse" in your kid's classroom, Haidt suggests a few very specific things you can do right now.
- Go "Dumb" on the Phone: If your kid needs a way to call you, give them a flip phone or a specialized "kid-safe" watch. No browser means no AI-cheating and no social media rabbit holes.
- The "Collective Action" Strategy: This is the most important one. You can't be the "mean parent" alone. Haidt suggests finding five other parents in your kid's friend group and making a pact: None of our kids get smartphones until 9th grade. It breaks the social pressure.
- Push for "Bell-to-Bell" Bans: If your school only bans phones "during instruction," it’s not working. Kids just check them in the bathroom or under the desk. Demand that phones stay in lockers the entire day so students actually have to talk to each other at lunch.
- Prioritize "Paper and Pen" Assignments: It sounds old-school, but in-class, handwritten essays are the only way to ensure a student is actually doing the cognitive heavy lifting. It’s "AI-proof" learning.
- Encourage "Risk" in the Real World: Give your kids tasks that have a chance of failure. Let them navigate the bus system, cook a meal, or build something without a YouTube tutorial or an AI guide.
The goal isn't to live in the 1950s. The goal is to make sure that when our kids finally do use AI, they are doing it as masters of the technology, not as its dependents. We want them to have "strong brains" that can think critically, feel deeply, and engage with the world without a screen as a mediator.
The "Great Rewiring" happened while we weren't looking. The "Great Unwiring" is going to take a lot more work.
Next Steps for Educators and Parents:
To dive deeper into the specific data Haidt uses, you should check out the After Babel Substack. He and his lead researcher, Zach Rausch, constantly update it with the latest studies on teen mental health and AI's role in the classroom. If you're ready to start a movement in your own school district, the Anxious Generation website has "Action Kits" tailored for PTA meetings and school boards.