What Kind of Future Are We Preparing Kids for as AI Changes School?
Last Tuesday, I stood by the Chromebook cart in room 214 and watched Mia stare at her screen like it had just told her the weather. Her essay on the Industrial Revolution was due at the end of the period. The paragraph in front of her was neat, confident, and completely empty—the kind of writing that says everything and nothing.
“Is this good?” she asked.
I read a sentence aloud. “‘The factory system transformed both labor and daily life in complex and lasting ways.’ What do you think?”
She shrugged. “I don’t know. It sounds right.”
That phrase has stayed with me. It sounds right. More and more, children can receive an answer that sounds polished, informed, and sure of itself before they have done the messy work of deciding what they believe. The issue is not simply cheating or shortcuts. The issue is that the uncomfortable middle of learning—the false starts, the bad sentences, the moments where you want to raise your hand and say you don’t get it—is now something many children try to skip.
I am a teacher, and I am also a parent, so I hear both sides of this. At school, teachers worry about essays that arrive in the same calm, generic voice. At home, parents wonder whether they are helping or falling behind when their child uses an automated tool for math, writing, or research. The larger question underneath all of it is simple: what kind of future are we preparing children for when machines can produce answers so quickly?
I don’t have a prediction about jobs in twenty years. I do know what children lose when they never practice being confused and then finding their way out.
In Mia’s case, I asked her to close the laptop for ten minutes. She groaned a little, which I took as a good sign. I gave her a blank piece of paper and said, “Write the worst thesis statement you can. Just tell me one thing you actually noticed.” She wrote that factories probably made life faster but also harder for families. It was not elegant. It was also hers.
Then she opened the laptop again, but the assignment had changed in her mind. She was no longer asking whether the machine’s answer was good. She was comparing it with what she thought.
That small shift matters. Children will need to work with automated tools, but they will also need to judge them. A generated answer may be confident and wrong. It may miss context, ignore people’s feelings, or smooth over the exact detail that makes a problem human. If a child cannot tell the difference between a convincing sentence and a true one, the tool does not help them think. It replaces thinking.
This is why I have changed some of my classroom routines. Students still use devices, but the first twenty minutes of writing often happen on paper. They write one claim by hand, turn to a partner, and say it out loud. If they use an automated tool later, they have to show me three places where they disagreed with it or changed it. I don’t expect them to reject every suggestion. I expect them to notice that they had a choice.
I also ask simpler questions in conferences now.
“Which sentence sounds most like you?”
“Where did you get stuck?”
“What would someone who was there actually care about?”
These questions slow work down, but that is the point. Learning is not always efficient. A child who only sees the final answer does not see the decisions hidden inside it.
At home, I see a version of the same thing with my son Leo, who is eleven. One evening, he was finishing math at the kitchen counter while pasta boiled. His screen showed a full solution, step by step. When I asked him to explain the second step, he pushed his glasses up and said, “It just did that.”
So we made a new rule. He tries the first two problems on the whiteboard before watching any video or asking for an automated explanation. If he uses help, he has to teach one step back to me. He hates the teaching part sometimes, but it works because it catches the difference between copying and understanding. I don’t need to know every feature of the tool. I only need to ask, “Can you show me where that number came from?”
That is a practical place for parents to start. You do not have to ban screens or become a technology expert. Put a piece of paper beside the laptop. Ask your child to draw the problem before typing it. Have them read their written answer aloud and circle one line they would change. Keep phones in a basket during the first fifteen minutes of homework, not because the devices are evil, but because beginning is hard and most shortcuts happen before momentum has a chance to build.
I also think about group work. Last month, a group used a tool to generate an entire project plan in thirty seconds. The plan looked organized, but none of them could explain who was doing what. They sat there, politely stuck, because the answer had skipped the human part: negotiating, noticing that Priya was good at drawing, that Daniel hated speaking but liked arranging slides, that one idea sounded great but would take three weeks.
That is another future skill children can only learn with other people. They need practice disagreeing without disappearing into a screen, noticing someone’s face, dividing work fairly, and fixing a plan when real life interrupts it. Machines can suggest a team structure. They cannot teach a child how to be a reliable teammate.
None of this means fear or refusal. The tools are here, and many of them are useful. A child can ask for a simpler explanation of a science concept, generate practice questions, or brainstorm ideas when stuck. But the order matters. Attempt first. Help second. Reflection last.
I still see Mia in the library sometimes. Her essays are not perfect. Last week she used a word she clearly did not need, and when I raised an eyebrow, she laughed and changed it. But she now starts more sentences with “I think” before she checks what the machine thinks. That may sound small. In ten or fifteen years, I think it may be the thing that matters most.
We are not preparing children for a future with no machines. We are preparing them for a future where machines handle more of the routine work, and humans are left with the harder questions: Is this fair? Does this make sense? Who is affected? What matters here? Those questions require judgment, and judgment grows slowly, through repeated small acts of trying, stumbling, and deciding.
The other afternoon, rain hit the classroom windows while students worked. Some wrote on paper, some typed, a couple argued quietly about evidence. One boy raised his hand and said, “The computer says one thing, but the source says another.”
I told him that was not a problem. That was the assignment.
He sighed, but he leaned in closer to the screen. I don’t know what his working life will look like. I only know he was practicing the right thing: not accepting the first confident answer, checking it against something real, and being willing to sit with the confusion long enough to form a thought of his own.
Please indicate: Thinking In Educating » What Kind of Future Are We Preparing Kids for as AI Changes School