Ms. Chen used AI to grade essays this year. It saved time. It kept the feedback consistent from the first paper to the last. She checked every word herself, sure of her oversight, but she kept her mouth shut about the algorithm’s help. There was no rule demanding transparency, and frankly, the line between her voice and the code’s hum was blurry.
Priya felt it immediately.
The comments lacked warmth. Ms. Chen’s usual honesty, that specific blend of critique and care, was gone. Something felt off. Next time, Priya didn’t submit her draft early. She didn’t say a thing, but the space between teacher and student grew cold.
This silence is everywhere now. It isn’t just a tech glitch, it is a relational fracture. Ms. Chen is drowning in work, trying to survive the load. Priya is a developing writer, and that human connection isn’t fluff, it is part of how she learns. Both are true. Both hurt.
Rupture doesn’t have to stick. But only if we name it.
The Ozempic Effect
Think of Fitz.
He is twenty-one. Known as the planner, the guy with the perfect spring break itineraries. This year he used AI to find hidden bars and weave in personal touches for his crew. They loved it. Best trip yet.
Then, during a chaotic archery-dodgeball game, a friend asked.
“How did you come up with that? Did you use ChatGpt?”
Fitz said no.
It was reflex, not strategy. Later that week, he wondered why he lied, especially when he prides himself on being straight-up.
He landed on an uncomfortable analogy: AI is the new Ozempic.
Not the drug, obviously. The cultural conversation. If you announce a strict diet and gym routine, we call that discipline. Respectable. Hard work. But if you take a GLP-1 shot to lose the same weight? The air changes. It gets loaded. We ask, Did they actually put in the effort? When a tool removes the struggle, we question the result’s value.
Fitz wanted to plan a good trip for his friends. Nothing malicious there. Yet his instinct was to hide the tool. Not because it was wrong, but because society hasn’t agreed on the rules yet. The norm vacuum is real, and in it, secrecy becomes the default survival mechanism.
We don’t need to pick sides on whether AI is “good” or “bad.” That’s a stale debate. The real goal is dragging young people out of the silent fog.
Silence Is Not a Policy
If left unchecked, silence is the norm. And for teens, that’s dangerous ground. Their identities are under construction, built brick by brick through peer feedback. They are hyper-aware of how others perceive them. One wrong move and you signal the wrong thing about who you are.
A norm vacuum breeds anxiety.
Young people are constantly scanning: Is what I did okay? Impressive? Something I need to scrub from my history? AI isn’t just a software update; it’s a social earthquake. And we have survived a similar tremor before.
Back in 2012. Smartphones went mainstream. Adults panicked, wrote bans, and handed control to platform developers who had one goal: engagement. Endless scrolling. Likes as social currency. The developers got rich, and youth mental health paid the toll.
We are repeating the mistake with generative AI. Harmful norms are forming right now, in the shadows, while adults blink slowly.
Some groups are trying. The Harvard Center for Digital Thracing offers classroom guides to ground values. The Rithm Project built a card game to spark chats about human connection. Good starts. They show we need space for kids to make sense of this on their own terms.
But guidelines don’t change culture. Peer networks do. The shift must happen through the fabric of actual social life, not top-down mandates.
Sensing, Reflecting, Nudging
Most institutions are set up to fix technical problems, not cultural ones. We default to handbooks, literacy tests, and fear-based policies. It misses the point entirely.
Changing culture requires three messy steps: Sensing. Reflecting. Nudging.
1. Sensing: Listen differently.
You can ask for feedback, or you can actually hear it. Surveys give data; they miss meaning. Why did the kid cheat? How did it feel to be caught? Tools like SenseMaker help here. Instead of rating AI usage on a scale of one to five, kids share anecdotes. Stories about avoiding it. Stories about feeling guilty. Follow-up questions unpack the emotion.
CIE uses this method for everything from phone bans to job training. It surfaces the invisible social codes beneath the hard data.
2. Reflecting: Make meaning together.
Data without discussion is just noise. An intergenerational team should review these patterns regularly. What is working? Where is the openness growing? Where is the hiding getting worse?
This isn’t a rubber stamp. Adults have to show up humbly, ready to be wrong. Youth and adults co-construct the reality.
3. Nudging: Shift the behavior.
A nudge is small, well-timed, and specific. Look at anti-smoking.
For years, ads showed black lungs and decayed teeth. Grim PSAs. They flopped because teens valued social autonomy over abstract long-term health risks. Florida’s Truth Campaign changed the angle. They asked teens why they smoked. It wasn’t ignorance. It was rebellion against authority. So they framed Big Tobacco as the oppressive manipulator. Refusing a cigarette became an act of defiance. Teen smoking plummeted from 28 percent to under 6.
Social currency works for AI too.
Right now, admitting to AI use feels like getting caught cheating. It signals laziness or fraud. A proper nudge reframes disclosure. It makes sharing tool-use an act of ownership, not shame.
Go back to Ms. Chen and Priya.
Imagine if they had sat down earlier. What if they arrived at a truth? Submitting work is trust. It requires a human witness.
They co-design a rule. First, peer review. Every essay gets human eyes, ensuring the writer feels seen. Second, Ms. Chen provides feedback, but she acknowledges the AI’s help in drafting the structure and syntax. It is curated. Open. Honest.
This agreement won’t live in a school handbook. Norms cannot be forced from the ceiling. They ripple. One person breaks the silence, then another follows, carrying the new expectation forward until it becomes simply what we do.
