For 150 years, every educational tool has been bent toward compliance and control. AI is the first one the system can't absorb. The students have the lever now.
What a brilliant essay. The writer has given us the picture of the education process that we all should have seen, having gone through it ourselves and having sent our children through it. The effects are seen everyday in the emptiness of personality in so many of our young people and even our peers. The writer makes a beautiful case for AI and how we can use it for good.
Fair challenge. In terms of the cognitive offloading concern, Gerlich's 2025 study documents it clearly, and I cite it in my own work. But the offloading isn't a property of the tool. It's a property of how the tool is used inside a system that never developed the capacity for independent judgment in the first place. If you hand AI to a student trained on rubrics and compliance, they offload. If you hand it to a student trained to evaluate, iterate, and critique, it becomes the most patient thinking partner they've ever had. The problem is that the second student barely exists, because the system wasn't designed to produce them. That's the argument in the piece — not that AI automatically amplifies thinking, but that it's the first tool that CAN, and the system's response to it reveals that amplifying thinking was never the priority.
Follow up question: are we talking about adolescents? Students aged 5 - 18? I don't think it's an issue with the system. It's about novice learners, how people learn, cognitive science...
The current research that I've seen is using undergrads, usually science/maths (and often at prestigious universities). I could well imagine that, being more at the (school) expert end of the spectrum, they could exhibit/use higher level skills with the requisite domain knowledge to find some useful purpose. And they're adults. Go to it.
So, what ages are you talking about, and what is there in cognitive science that supports your premise?
Good questions. I'm talking about all ages — the argument is about what the system produces, not about developmental readiness. A 12-year-old trained to evaluate her own work and iterate on critique will use AI differently than a 22-year-old who spent twelve years performing against rubrics. The cognitive science on novice learners is sound, but it describes how people acquire knowledge — not how they develop judgment. The system optimized for the first and abandoned the second. That's a design choice, not a developmental inevitability.
All that said, the piece is rhetoric, not research — it's arguing that the system's priorities are visible in its behavior. Ultimately, young people don't need peer-reviewed permission to pick up a tool and use it in ways the system didn't anticipate. They just need the system to stop getting in the way.
What a brilliant essay. The writer has given us the picture of the education process that we all should have seen, having gone through it ourselves and having sent our children through it. The effects are seen everyday in the emptiness of personality in so many of our young people and even our peers. The writer makes a beautiful case for AI and how we can use it for good.
Interesting... where is the evidence that, for students, AI "amplifies thinking"?
There seems to be more of an indication that it provides an opportunity for cognitive offloading.
Fair challenge. In terms of the cognitive offloading concern, Gerlich's 2025 study documents it clearly, and I cite it in my own work. But the offloading isn't a property of the tool. It's a property of how the tool is used inside a system that never developed the capacity for independent judgment in the first place. If you hand AI to a student trained on rubrics and compliance, they offload. If you hand it to a student trained to evaluate, iterate, and critique, it becomes the most patient thinking partner they've ever had. The problem is that the second student barely exists, because the system wasn't designed to produce them. That's the argument in the piece — not that AI automatically amplifies thinking, but that it's the first tool that CAN, and the system's response to it reveals that amplifying thinking was never the priority.
Follow up question: are we talking about adolescents? Students aged 5 - 18? I don't think it's an issue with the system. It's about novice learners, how people learn, cognitive science...
The current research that I've seen is using undergrads, usually science/maths (and often at prestigious universities). I could well imagine that, being more at the (school) expert end of the spectrum, they could exhibit/use higher level skills with the requisite domain knowledge to find some useful purpose. And they're adults. Go to it.
So, what ages are you talking about, and what is there in cognitive science that supports your premise?
Good questions. I'm talking about all ages — the argument is about what the system produces, not about developmental readiness. A 12-year-old trained to evaluate her own work and iterate on critique will use AI differently than a 22-year-old who spent twelve years performing against rubrics. The cognitive science on novice learners is sound, but it describes how people acquire knowledge — not how they develop judgment. The system optimized for the first and abandoned the second. That's a design choice, not a developmental inevitability.
All that said, the piece is rhetoric, not research — it's arguing that the system's priorities are visible in its behavior. Ultimately, young people don't need peer-reviewed permission to pick up a tool and use it in ways the system didn't anticipate. They just need the system to stop getting in the way.