What changes when 4th graders use AI differently
It starts with a single word a 4th grader couldn't figure out.
Hey there,
Picture independent reading time in a 4th grade classroom. A student hits a word they’ve never seen — “desolate” — and pauses. They’re reading this:
“The barn stood alone at the edge of the field. The land around it was desolate — nothing moved, nothing grew, and the wind made a sound like something that had given up.”
This is a podcast overview of the entire newsletter in case that you prefer to listen to it in this format. This podcast overview was created with NotebookLM.
They re-read the sentence. Their brain starts working: maybe it means empty? Maybe lonely? It’s a reasonable guess, and honestly, it’s the right instinct. They scanned the emotional tone of the passage and connected the clues. That's the instinct every reading teacher hopes to see. But here’s what happens next: nothing. If they raise their hand, the teacher gives them the definition. If they don’t, they skip the word and keep reading with a gap in their understanding. Either way, the thinking they just did vanishes. No one confirms it was good. No one builds on it. The student learns, slowly and quietly, not to trust their own reasoning.
Today’s post is sponsored By: Project Pals
This happens thousands of times across a school year, and it has nothing to do with AI. AI didn’t create this problem, but used the right way, it might be the first tool that can actually fix it. Most school conversations about AI circle back to cheating and shortcuts, which is the wrong starting point. Students already have access to AI. The real question is whether teachers have a framework that channels that access into genuine thinking — or just leaves it to chance.
That’s what today’s newsletter is about:
Why the pre-AI classroom was already losing students’ best thinking
How one 4th grade lesson plan changes that dynamic entirely
What it looks like when AI asks the questions instead of answering them
Let’s get into it.
If you’re an educator trying to figure out how to bring AI into your classroom without watching students use it as a shortcut, here are five resources worth your time:
Weekly resource list:
UNESCO’s AI Competency Framework for Teachers — The clearest global benchmark for what AI literacy actually looks like for educators. A strong foundation before any classroom implementation.
ISTE’s Hands-On AI Projects for the Classroom — Practical, subject-specific project ideas from one of the most trusted names in ed-tech. Pairs well with the approach covered below.
The Socratic Method: Fostering Critical Thinking This article explores how the Socratic Method can be used to promote critical thinking in classroom discussions.
AI Literacy Lessons for Grades 6–12 — Quick, grab-and-go lessons to help your students think critically about AI and its impact.
Using AI as a Thought Partner, Not a Shortcut— A grounded piece on keeping students in the driver’s seat when AI is in the room.
The feedback loop students never had
The problem in that reading moment was never a lack of access to information. A dictionary was usually somewhere nearby. The problem was that students had no feedback loop for their own reasoning. They’d done good thinking and had no way to know it, so uncertainty felt identical to being wrong.
Here’s what the 4th grade AI integration guide does with that moment. A student hits the word “desolate” and instead of Googling the definition, they open ChatGPT and type:
“I’m trying to figure out what ‘desolate’ means. I think it might mean alone or empty because nothing moved or grew. Please don’t tell me the definition — ask me questions to help me figure it out.”
ChatGPT responds:
“You’re already noticing an important clue. What details in the sentence make the place feel empty or lifeless? Which words seem strongest to you — ‘stood alone,’ ‘nothing moved,’ ‘nothing grew,’ or ‘the wind made a sound like something that had given up’? What feeling do all of those clues create when you put them together?”
The student works through it and lands on empty, lifeless, abandoned. ChatGPT pushes further:
“If a place is alone but still full of plants, animals, and activity, would it feel the same way as this barn?”
By the end, the student has arrived at the meaning themselves — not because someone told them, but because they were questioned until their own reasoning got them there. That’s the whole method, and it transfers across every subject. In math, students describe word problems to AI in their own words, and AI asks clarifying questions to help them figure out what the problem is actually asking. In science, AI prompts students to state what they already know before forming a hypothesis. In social studies, students interview AI as a historical “witness” and then fact-check those answers against classroom sources, which builds sourcing habits most curricula spend years trying to develop.
For teachers, the practical benefit is that thinking becomes visible. When students submit their AI conversation logs alongside their final work, you’re grading the reasoning, not just the finished product. You can see where they got stuck, what questions moved them, and whether they pushed back or just went along. That’s assessment information a polished essay almost never gives you, and it changes what feedback you can offer. None of this needs special software or training — the framework works with whatever AI tools your students already have.
Here’s what you walked away with today:
The pre-AI classroom was already losing students’ best thinking at the moment of independent work — there was no feedback loop for self-generated reasoning
This approach closes that loop by instructing AI to ask questions instead of giving answers
It works across ELA, math, science, and social studies, and gives teachers a window into student thinking that final products alone never show
The 4th grade AI integration guide already has the prompts written for you. There’s an opening prompt, a prompt for when students get stuck, and a reflection prompt for after the conversation ends. Pick one lesson where students typically hit a wall, pull the relevant prompt from the guide, and see what the conversation logs tell you about how your students actually think. The framework is already there — you just have to use it.
PS...If you’re enjoying Master AI For Teaching Success, please consider referring this edition to a friend. They’ll get access to our growing library of AI prompts and templates, plus our exclusive “Popular AI Tools Integration Guide For Teachers“ implementation guide.