Dr Phillippa Hardman shares some fascinating ways learners are using GenAi to learn. As I read her excellent piece, I wondered, “What might prompts look like for the various ways learners use GenAi AI? What prompts could illustrate these uses?”

So I asked ChatGPT and then layered in Hattie and instructional coaching into the response. I will use this as part of the knowledge base when I begin planning some coaching sessions in 2026. I find this fascinating…

---

Four Learner Uses of AI, Mapped to Instructional Impact

What learners are doing with AI aligns closely with what we already know about effective instruction. The difference is that learners are now accessing these supports on demand, without waiting for formal design or delivery.

1. AI as Default Learning Infrastructure

Aligned Hattie strategies

• Teacher clarity
• Prior knowledge activation
• Advance organizers

Learners routinely use AI to orient themselves: clarifying goals, defining terms, and establishing baseline understanding before deeper work begins.

Illustrative prompts

• “Explain the core idea from today’s lesson in simple language.”
• “What background knowledge do I need before this topic makes sense?”
• “Give me a short overview so I know what matters most.”

Why this matters instructionally

Teacher clarity has one of the strongest effects on learning, yet it is often uneven or delayed. Learners are using AI to self-generate clarity before confusion compounds.

Coaching implication

Rather than coaching teachers to re-explain content endlessly, instructional coaching can focus on:

Designing lessons where clarity is surfaced early Embedding prompts that explicitly ask learners to generate their own “clarity check” using AI, then compare it to the teacher’s intent Coaching teachers to analyze where learner-generated clarity diverges from instructional goals

2. AI as a Tutor and Thinking Partner

Aligned Hattie strategies

• Worked examples
• Scaffolding
• Feedback (process-level, not task-level)

Learners consistently ask AI to explain thinking, model processes, and guide them through reasoning.

Illustrative prompts

• “Walk me through this step by step and explain why each step works.”
• “Show me an example, then give me a similar one to try.”
• “Ask me questions that help me figure this out myself.”

Why this matters instructionally

These prompts map directly onto how novices build understanding. Learners are recreating effective tutoring patterns when those patterns are not reliably available.

Coaching implication

Instructional coaching can shift from:

“How do I explain this better?” to “Where do learners need modeling, questioning, or worked examples built into the task?”

Coaches can help teachers:

• Design tasks where AI-supported scaffolding is intentional, not hidden
• Require learners to document how AI-supported explanations changed their understanding
• Use student prompts as evidence of where scaffolding is insufficient or misaligned

3. AI During Focused, Deliberate Work

Aligned Hattie strategies

• Metacognitive strategies
• Self-regulation
• Feedback timing

Learners turn to AI mid-task, not after giving up. They use it to test reasoning, request hints, and validate progress.

Illustrative prompts

• “I am stuck here. Give me one hint.”
• “Where does my reasoning first stop making sense?”
• “What is a common mistake I might be making?”

Why this matters instructionally

Metacognition and self-regulation are high-impact influences, yet they are rarely taught explicitly. Learners are now practicing them through AI-supported self-checks.

Coaching implication

Instructional coaching can focus on:

• Designing tasks that pause for prediction, verification, and revision
• Coaching teachers to ask for “decision trails” instead of polished products
• Using AI-supported attempts as artifacts to diagnose shared misconceptions

This repositions feedback from something teachers deliver to something learners actively seek and evaluate.

4. AI as a Repair Tool for Instructional Gaps

Aligned Hattie strategies

Deliberate practice Success criteria Challenge and practice balance

When instruction lacks accessible explanations or practice sequences, learners use AI to rebuild them.

Illustrative prompts

• “Create practice problems that increase in difficulty.”
• “Quiz me and adapt based on my answers.”
• “Give me a checklist for how to approach this kind of task.”

Why this matters instructionally

Learners are not asking for easier work. They are asking for structured practice and clearer success criteria so they can persist.

Coaching implication

This is powerful diagnostic data. Coaches can help teachers:

• Identify where learners are compensating for missing design elements
• Replace generic assignments with practice sequences that surface thinking
• Use learner-generated AI scaffolds as feedback on instructional design quality

Weak design is no longer invisible. AI makes it visible at scale.

What This Means for Instructional Coaching Overall

AI does not replace instructional design or coaching. It raises the bar.

As learners increasingly access:

• Clarity on demand
• Process-level feedback
• Worked examples
• Metacognitive prompts

The educator’s role shifts toward:

• Designing tasks worth doing even with AI present
• Diagnosing patterns of misunderstanding
• Coaching learners in how to think with support, not without it

Instructional coaching, in this context, becomes less about compliance or tool adoption and more about aligning learning design with how learners already learn.

Learners have not waited for permission.

They have built their own support systems.

The question now is whether instruction and coaching are designed to meet them there.