Every slide is posted. Every lecture is on the class site. Half their teachers hand out the outline before the unit even starts. Your kid walks around with a more complete record of the school day than you ever had, and remembers less of it than you did.
That gap is the whole story, and almost nobody is telling it straight. The conversation about kids and note-taking has collapsed into two tired camps: they're lazy, or they're on their phones. Now there's a third panic layered on top, that AI is going to finish the job and hollow out their brains completely. All three miss the same thing. To see it, you have to notice that note-taking was always doing two separate jobs, and we only ever paid attention to one of them.
The record was never the point
Note-taking records information. It also forces you to decide, in real time, what is worth recording. Those are different jobs. The first one produces a document. The second one is thinking.
For most of the history of school, the two came bundled. You couldn't get the record without doing the deciding, because the only way to capture a lecture was to run it through your own head fast enough to write down the part that mattered. The deciding was invisible. It felt like the writing. So when we told kids to "take good notes," we thought we were asking for a document, and we were actually asking for the thinking that produced it.
Then technology unbundled the two. Slides get posted. Lectures get captured. The learning management system holds a cleaner, more complete record than any fourteen-year-old could produce with a pen. The document is handled. What got quietly deleted was the only reason a kid ever had to do the deciding.
So a kid looks at the posted slides and reaches the obvious conclusion. Why would I write down what's already written down? They are not wrong about the record. They are missing what the notes were actually for. And they're missing it because we never told them, because we didn't know it either.
Does note-taking still matter when AI takes better notes than any kid?
Here's the question every parent and student is quietly asking now, even if they don't say it out loud. If the slides are posted and AI can summarize the entire chapter in four bullet points, why take notes at all?
The answer is yes, it still matters, and the reason is the part that survives when you take the record away. The value was never the four bullet points. It was the work of getting to them: reading the chapter, deciding which four points out of forty actually carry the idea, and putting them in your own words. That work has a name in the research. It's called encoding, and it is most of what learning actually is.
What the research actually shows
This isn't a hunch. In one of the more quietly damning studies in education, researchers Pam Mueller and Daniel Oppenheimer had students take notes on lectures, some by hand, some on laptops. The laptop group recorded more words. They captured more of the lecture, a better record by any obvious measure. And they scored worse on the questions that asked them to actually understand the material.
More words, worse learning, from the same behavior. The laptop let them transcribe without deciding, and transcription without deciding is a photocopier with a heartbeat. The students who learned were the ones forced to compress, because compression is the act of figuring out what matters. That's the finding that should end the debate about whether the record is the point. A better record produced worse understanding.
There's a second finding that matters just as much, and it's the one that lets schools off the hook far too easily. Note-taking is teachable. When students get explicit instruction in how to do it, the quality of their notes and how much they remember both go up. Which means the kids who never learned it didn't fail at something they were taught. They were handed an instruction, "take good notes," and left to reverse-engineer a skill nobody ever explained. Some figured it out by accident. Most didn't.
AI didn't cause this. It revealed it.
Now put AI on top of a skill that school already stopped requiring, and you get the numbers everyone is alarmed about. Between May and December of 2025, the share of students using AI for homework rose from 48 percent to 62 percent, driven mostly by middle and high schoolers. By December, the share of students who believe AI is harming their critical thinking had climbed to 67 percent, up from 54 percent at the start of the year.
Sit with that second number. The kids can feel the skill not forming. Two out of three of them will tell you the tool is costing them something, and they reach for it anyway, because the friction is gone and the assignment is due. They just can't name what they're losing, because nobody ever named it for them.
AI didn't create this problem. It removed the last bit of friction that was still, by accident, making a few kids do the encoding by hand. The kid who uses AI to write every essay doesn't have a discipline problem. They have a capacity that was never built, the same one they were missing before AI existed. AI just made it impossible to hide.
The wrong question about AI in school
Which is why the entire debate about AI in schools is stuck on the wrong verb. Every district meeting, every op-ed, every detection tool is aimed at the same question: how do we stop students from using AI to produce the work. The output is not the problem. The input that never happens is the problem.
The better question is the one almost nobody is asking. What skill would make AI a tool in a kid's hands instead of a substitute for one? Because a student who can encode, who can read something and decide what matters and put it in their own words, uses AI to go faster and check their thinking. A student who can't uses it to skip the part where the learning happens. Same tool. Opposite outcome. And the thing that determines which kid you get is something no phone ban and no detection software will ever build.
What encoding actually looks like
Here's the part you can use tonight, and it's simpler than the panic suggests.
Encoding is translation. Instead of writing down what the teacher said, you write down what it meant. The teacher says, "Photosynthesis is the process by which plants convert light energy into chemical energy stored in glucose." You don't copy that. You write, "Plants use sunlight to make sugar for energy." Same information. But you had to understand it to shrink it, and because you understood it to shrink it, you'll remember it.
Underneath that is a structure that turns a page of fragments into something usable. Every piece of information a kid captures belongs to one of three layers. The main idea, the one thing they'd say if you asked what the class was about and gave them ten seconds. The supporting details, the examples and evidence that make the main idea hold. And the open questions, the parts that didn't click, written down on purpose instead of quietly skipped. Notes that carry all three are notes a kid can actually study from. Notes that are just a stream of whatever came at them are three pages that get thrown out in June.
None of this requires a fancy method or an app. It requires the kid to do the deciding that the posted slides now do for them. That's the whole skill. It's small, it's teachable, and it's the difference between a kid who owns their learning and a kid who is a passive channel for information moving past them.
The part that's actually yours
If you've spent the last few years becoming the parent who became the external memory, reminding, checking, catching the thing before it dropped, this is the same problem wearing a different coat. The goal was never for you to hold the system. The goal is a kid who can run it. Encoding is one of the pieces they build to get there, and it's one of the eight weeks of skills in the Field Guide to High School, which walks a rising ninth grader through building the whole toolkit before the year that assumes they already have it.
But you don't need the book to start. Tonight, when your kid says the teacher posts everything, you can tell them the truth nobody told you. The slides were never the point. The point was the two seconds where you decide what matters. That decision is the skill. Everything AI can do for them is downstream of whether they can still make it themselves.
Sources:
- Pam A. Mueller and Daniel M. Oppenheimer, "The Pen Is Mightier Than the Keyboard: Advantages of Longhand Over Laptop Note Taking," Psychological Science, 2014
- Joseph R. Boyle, research on explicit strategy instruction and student note-taking, 2013, summarized in "Note-Taking: A Research Roundup," Cult of Pedagogy
- RAND Corporation, "More Students Use AI for Homework, and More Believe It Harms Critical Thinking: Selected Findings from the American Youth Panel" (RR-A4742-1), 2026