Claude Academy
Loading

Creating high quality AI outputs

Lesson 625 min
In this lessonBy the end, you’ll be able to
  • Use Description skills to guide AI in creating differentiated instructional materials
  • Use Discernment skills to evaluate AI-generated materials for accuracy, reading level, and alignment to your learning objectives

Creating high quality AI outputs in your teaching practice

Loading

Creating high quality AI outputs · 10 min

SummaryTranscript

This video walks through the Description-Discernment loop. You'll watch Ms. Okafor, a 7th grade science teacher, use AI to draft a differentiated informational text for a wide range of readers, then apply her content expertise to evaluate and revise what AI produces. You'll see how specific context, iterative feedback, and pedagogical judgment turn a generic AI draft into something that actually fits a real classroom.

The Description ↔ Discernment loop

Fluent AI use happens in a tight loop: you describe what you want, the model responds, you discern what's useful, and you describe again, sharper this time.

DescriptionCommunicate the vision so the model can act on it.
DiscernmentJudge what came back, and feed that judgment forward.
Describe what good looks like.Clear, specific prompting that turns your vision into terms the model can act on.
PromptInner LoopRefine
Judge what came back.Judge what comes back—quality, relevance, bias—so your next prompt is sharper.
Quick check

Up front, you tell AI your students read below grade level and need key vocabulary defined in context. Which move is that?

Key takeaways

  • Effective Description provides specific context. The more you share about you, your students' needs, and your teaching habits, the better.
  • Discernment isn't optional. Evaluate outputs when accuracy matters—flag claims to verify or ask for citations.
  • The loop is iterative. Your first prompt rarely nails it; refine using what each response teaches you.
  • AI accelerates, but doesn't replace expertise. You're still the decision-maker.
  • Upload past materials. Your unit plans, rubrics, or strong examples help AI match your voice and rigor.

Exercises

Differentiated materials challenge

Apply Description and Discernment to create a piece of differentiated instructional material for a real upcoming lesson, just like Ms. Okafor did.

Part I: Describe

  1. Pick something you'll actually use and need—some ideas: a read-aloud with discussion questions, a leveled text, a math station, a visual support, a practice set, a center activity.
  2. Draft a Description prompt that names the standard or learning target it supports, the range of learners in your room, at least one specific accommodation need, and what students will do with it.

Part II: Discern

  1. Share your prompt with an AI tool and review the output against Ms. Okafor's Discernment checklist:
    • Is the content accurate for your subject area?
    • Did it land at the right level for your learners, with key vocabulary or concepts introduced and supported?
    • Does it support what students will do with it and where the lesson needs to go?
    • Would anything land badly with your specific students?
  2. Write one concrete revision prompt (not "make it better," but e.g. "shorten the steps in the center directions and add a picture cue for each one"), submit it, and compare the revision to the original.

Stretch goal: Build your revised material into a student-facing-quality product (lesson guide, center, project, worksheet) with clear directions, objectives, scaffolds, and a check for understanding—running the Description-Discernment loop until it's something you'd stand behind.

Lesson reflection

  • How did giving AI rich context about your students and goals change its output, compared to a generic request?
  • What's one discernment check you'll always run before putting an AI-made material in front of students?

What's next

In the next lesson, we'll put Delegation and Diligence into practice with a data-analysis scenario. This includes how to decide what AI should touch, how to protect student information, and how to validate results before you trust them.