Posted November 19, 2025

By: Olivia Odileke

The Surface Level of AI Use

Let's be honest—most of us are still scratching the surface when it comes to using AI in education.

We've asked it to write a few emails. Generate a rubric. Brainstorm a lesson hook or discussion questions. Maybe we've even had it draft parent communication or create a template we can reuse.

And all of that is useful. But it's not transformative.

Because here's what we're not doing: we're not inviting AI into the messy, complex, high-stakes thinking that actually drives educational change.

How often do we use AI to help us understand why a student is unmotivated—not just what activity might engage them, but what's underneath the disengagement? How often do we ask it to help us redesign a team workflow that's creating bottlenecks, or to interpret patterns in assessment data that we're too exhausted to see clearly?

We're using AI like a vending machine: insert request, receive output, move on.

But what if we could use it like a thinking partner—someone who sits beside us in the complexity, asks clarifying questions, surfaces blind spots, and helps us reason our way to better decisions?

That's not a hypothetical future. That's available right now. We just haven't learned to access it yet.

The Shift: From Tool to Thought Partner

We're entering an era where AI can do more than just produce content; it can help us reason.

Not in the way a human reasons—AI doesn't have intuition, doesn't understand the cultural nuances of your school, doesn't know the history between two colleagues who can't seem to work together. But it can do something humans struggle with when we're overwhelmed: it can simulate options, weigh trade-offs, model consequences, and identify root causes we might overlook when we're too close to the problem.

Think of AI as your co-analyst, co-designer, and co-reflector—a partner that helps you make sense of complex decisions by bringing structure to ambiguity.

But here's the critical shift: you have to prompt it differently.

Most of us have been trained to use tools—which means we issue commands and expect compliance. "Create this." "Generate that." "Write this in three bullet points."

But problem-solving isn't transactional. It's conversational. It's iterative. It requires context, clarification, and back-and-forth refinement.

The key is learning to prompt AI like you would mentor a colleague:

"Here's what's happening. Here's what I've tried. Here's what I'm noticing. Here's what I'm wondering. Help me think through the next step."

That's not a command.
That's collaboration.

And when you approach AI with that mindset, something remarkable happens: you stop asking it to think for you, and you start using it to think with you. Your expertise stays at the center. Your judgment drives the decision. AI just helps you see more clearly.

A Real-World Example: Turning Data into Action

Let me show you what this looks like in practice.

Last month, I worked with a team of instructional coaches who were struggling to interpret interim assessment data. They knew their students were underperforming in critical areas—reading comprehension for middle schoolers, particularly around making inferences—but the "why" was buried under spreadsheets full of numbers that all looked equally concerning.

They'd done what most teams do: sorted by proficiency levels, color-coded the struggling students, and scheduled intervention blocks. But they still didn't understand the pattern. Why were these particular students struggling? Was it a vocabulary issue? A lack of background knowledge? Insufficient modeling during instruction? Something about the assessment design itself?

They had data. They didn't have insight.

So we tried something different. We uploaded a sample of anonymized data into ChatGPT and prompted it:

"What learning patterns do you see here? What hypotheses could explain these trends? What questions should we be asking that we haven't asked yet?"

Within minutes, AI grouped the data in ways they hadn't considered—clustering students not just by proficiency level, but by error patterns. It surfaced possible misconceptions (like students consistently choosing literal answers when inference was required). And it generated probing questions for their next PLC meeting: "Are we teaching students to distinguish between 'what the text says' and 'what the text means'? How are we modeling the invisible thinking that proficient readers do automatically?"

Here's what struck me: the insight wasn't the chart it produced. It was the thinking it provoked.

The coaches left that session with something they hadn't had before—not answers, but clarity. And with that clarity came curiosity. They were ready to test hypotheses rather than chase symptoms. Ready to try targeted instructional moves instead of generic interventions.

That's the power of AI as a thought partner. It didn't solve their problem. It helped them see their problem more clearly—which is often the only thing standing between us and the solution.

Why Educators Haven't Reached This Level Yet

If AI can do this, why aren't more educators using it this way?

It's not because teachers lack curiosity or intelligence. It's because we're navigating three very real barriers:

1. Lack of Time to Explore Deeply

The same time scarcity that makes AI appealing (we need help now) also prevents us from learning to use it well. It's faster to ask for an email template than to engage in reflective inquiry. It's easier to grab a quick output than to iterate through a problem-solving conversation.

But this is a false economy. Five minutes spent generating a surface-level solution might save time today—but it won't build your capacity to solve the next ten similar problems on your own.

2. Fear of Imperfection

We've been conditioned to see tools as either working or not working. If the AI doesn't "get it right" on the first try, we assume it's not useful and we move on.

But problem-solving is inherently iterative. You test an idea, see what happens, adjust, and try again. That's not AI failing—that's exactly how thinking works. And that's where AI actually shines: in the iteration, the refinement, the back-and-forth that helps you sharpen your strategy.

3. We've Been Trained to Use Tools, Not to Co-Think

This is the deepest barrier: education has conditioned us to look for right answers, not collaborative inquiry.

From the time we were students ourselves, we learned that thinking has a destination. You solve the problem. You arrive at the answer. You close the loop. But the most complex challenges in education—equity gaps, student motivation, teacher retention, community trust—don't have neat answers. They have tensions to navigate and trade-offs to weigh.

AI invites us to unlearn that scarcity of thinking. It asks us to treat problem-solving as an evolving process, not a finite task. And for educators who've spent careers optimizing for efficiency, that's a profound shift.

From Problem to Partnership: A Framework for Real-Time Thinking

So how do you actually make this shift? How do you move from transactional prompting to collaborative thinking?

Here's a framework I use—and teach—for turning AI into your real-time problem-solving partner:

1. Describe the Problem, Not the Task

Instead of asking AI to produce something ("Write a letter to parents"), describe the challenge you're trying to solve ("Help me explain a sensitive schedule change to families in a way that builds trust rather than triggering defensiveness").

Why this matters: When you frame the task, AI gives you a template. When you frame the problem, AI helps you think strategically about what needs to happen and why.

Example:

• Weak prompt: "Create a behavior intervention plan"
• Strong prompt: "I have a 4th grader who's been acting out during transitions. She's academically capable but seems overwhelmed when routines change. Help me think through what might be driving this behavior and what small adjustments I could test first."

2. Provide Context

Include grade level, school setting, constraints, and relevant background. The more real you are with AI, the more relevant it becomes.

Context isn't just demographics—it's also your thinking so far: "I've already tried giving her advance notice, and it helped slightly but not consistently."

Why this matters: AI can't read your mind, but it can work with your context. The richer the input, the sharper the output.

3. Ask for Reasoning, Not Just Results

Prompt AI with questions like:

• "Walk me through your thinking on this."
• "What are three possible root causes I should consider?"
• "What am I not seeing that might be important?"

Why this matters: This turns AI into a thinking coach, not just a content generator. You learn not just what to do, but how to think about similar problems in the future.

4. Test and Refine

Treat each AI exchange like a hypothesis. Try the strategy. Reflect on what happened. Come back and say: "I tried that approach with my student, and here's what I noticed. What should I be paying attention to next?"

This is where the magic happens. AI becomes a longitudinal thinking partner—helping you track your own learning across multiple problem-solving cycles.

The Human Edge: What AI Can't Replace

Let me be clear about something important: AI is powerful, but it doesn't replace human intuition.

AI can model possibilities, simulate scenarios, and surface patterns in data. But it can't measure what matters most in education: the quality of relationships, the health of your school culture, the depth of belief you have in your students' potential.

AI can tell you what might work. Only you can decide what's worth trying.

That's why we need both—not AI instead of human judgment, but AI in service of it.

AI brings method and structure when we're overwhelmed.
We bring meaning, values, and relational wisdom.

Together, they help us lead with calm, clarity, and curiosity—the foundation of every transformation that lasts.

Your Invitation: Shift from Production to Problem-Solving

This week, I challenge you to use AI not to produce something, but to solve something.

Start with one real challenge you're facing:

• A student who's disengaged despite your best efforts
• A team dynamic that's creating friction
• A scheduling puzzle that feels impossible
• An instructional strategy that's not landing the way you hoped

Ask AI to help you think it through—not to give you the answer, but to help you see the problem more clearly.

Then pause afterward and reflect: How did this conversation change my approach? What did I notice about my own thinking? What would I do differently next time?

You'll start to see what I see—not a tool that replaces us, but a partner that refines how we think, decide, and lead.

Because AI isn't here to replace our thinking.
It's here to elevate it.

And when we use it that way—as a thought partner, not a shortcut—we don't just get better outputs.
We become better educators.