AI Learning Modules: Tools and Decision Systems
University of Arizona | College of Education
Module four of a four-module program helping educators build foundational AI literacy, establish ethical guardrails, master prompt design, and navigate AI tools with confidence.
Created by Austin Ross, M.Ed., MA
Access U of A GenAI

Module Overview
This module guides University of Arizona educator preparation program instructors through key ideas related to AI tool use, assignment-level decision-making, and prompting in educational settings. Three core topics are covered in sequence.
01
Key Terms
Foundational vocabulary and concepts for understanding AI in education.
02
Tools Landscape
Expectations and practices for keeping educators actively involved in AI-assisted decisions.
03
Stoplight System
Protecting student privacy and understanding data compliance when using AI tools.
04
Prompt Frameworks
Foundational strategies for writing stronger prompts that lead to more useful, targeted, and instructionally relevant outputs.
05
Video Library
This section includes optional tutorials and resource links for anyone who wants to spend more time exploring the tools beyond the main module. These videos can help faculty and instructors get more familiar with platforms like MagicSchool, Canva Code for Me, and other AI tools that may support planning, productivity, and instructional design.
Tools and Decisions Welcome


Glossary of Key Terms
Before diving into AI prompting, it helps to have a shared vocabulary. The terms below span instructional design, assessment, and AI policy — all of which inform how we plan, prompt, and verify AI-generated lesson materials.
Instructional Design
  • Anticipatory Set — A short activity at the start of a lesson to capture attention and activate prior knowledge.
  • Scaffolding — Temporary supports that help students learn new concepts before gradually removing assistance.
  • Differentiation — Adjusting instruction, materials, or assessments to meet varied student needs and readiness levels.
  • Closure — A short end-of-lesson activity that summarizes learning and checks for understanding.
Assessment & Engagement
  • Engagement Strategies — Instructional approaches that actively involve students and increase participation.
  • Formative Assessment — Ongoing checks for understanding used during instruction to guide teaching decisions.
  • Summative Assessment — Evaluations at the end of a lesson or unit to measure student learning.
AI & Policy
  • Artificial Intelligence (AI) — Computer systems that perform tasks typically requiring human intelligence, such as generating text or analyzing patterns.
  • Prompt Frameworks — Structured approaches for writing prompts that help AI generate clearer, more accurate responses.
  • FERPA — The Family Educational Rights and Privacy Act, protecting the privacy of student education records.
  • COPPA — The Children's Online Privacy Protection Act, protecting personal information of children under 13 online.
Tools and Decision Systems
AI tools are not all built for the same kind of work, and they should not all be treated the same way in teaching, planning, or professional tasks.
This page is meant to help faculty think more intentionally about: which AI tools make sense for which tasks, how to make clearer decisions about acceptable use, how to communicate those expectations to students, and how better prompting can lead to better outputs.

Rather than asking, "What is the best AI tool?" a better question is, "What tool makes the most sense for this task, in this context, with these boundaries?"
Start with U of A GenAI
For University of Arizona faculty and students, chat.ai2s.org is one of the most important starting points for AI use.
This matters because it gives our campus community a shared space to explore AI tools in a way that is more aligned with institutional privacy expectations and educational use. It also allows users to compare different models in one place instead of jumping randomly between public tools.

That does not mean anything can be pasted into it. Faculty still need to use judgment and avoid entering restricted, highly sensitive, or regulated information.
The Tools Landscape
One of the biggest mistakes people make with AI is treating every tool like it does the same thing.
Some tools are better for:
Brainstorming
Summarizing
Drafting
Revising
Multimodal work
Classroom resource generation
Teacher workflow support
Questions worth asking:
What am I actually trying to do?
Do I need speed, depth, structure, or creativity?
Is this a low-stakes productivity task or a higher-stakes instructional task?
Would a general AI tool help, or would an education-specific tool make more sense?
When we talk about a tools landscape, we are really talking about matching the tool to the task.
Featured Tools You May See
Inside the University of Arizona AI environment, you may see several model options. You do not need to memorize every name or technical detail. What matters more is getting a rough sense of what different tools may be better at.
Fast Drafting & Summarizing
Some models are better for quick drafts and condensing information efficiently.
Complex Reasoning
Some are stronger at more complex reasoning and multi-step problem solving.
Multimodal Tasks
Some are built to handle multimodal tasks involving text, images, and more.
Open-Model Options
Some are open-model options worth exploring for flexibility and transparency.
Educator-Facing Tools
Some are educator-facing tools built around teaching workflows — including MagicSchool, designed specifically for educational use: planning, differentiation, formative assessment ideas, and instructional resource creation.

The goal is not to become a model expert. The goal is to become more intentional.
MagicSchool and Teacher Prep
MagicSchool is especially relevant in teacher education because it is built around tasks that feel familiar to educators.
Faculty or teacher-candidates might use it to:
  • Brainstorm lesson hooks
  • Create draft exemplars
  • Generate differentiation ideas
  • Build quick checks for understanding
  • Create discussion prompts or resource supports
Outputs still need to be reviewed for:
  • Alignment
  • Accessibility
  • Developmental appropriateness
  • Cultural responsiveness
  • Actual usefulness in context

That said, educator-facing does not automatically mean instructionally strong. AI can save time, but it still needs human judgment. Refer back to human-in-the-loop in Webpage 2.
The Stoplight System
A stoplight system is one of the clearest ways to communicate expectations around AI use.
🔴 Red Light
No AI use permitted. This makes sense when the task is meant to measure a student's own writing, thinking, analysis, or disciplinary skill without AI support.
🟡 Yellow Light
Limited AI use permitted. This makes sense when AI can support brainstorming, organizing, outlining, or revising, but the student still needs to do the actual intellectual work and remain accountable for the final product.
🟢 Green Light
AI use encouraged. This makes sense when the goal is AI literacy, critique, experimentation, workflow support, or professional application.
Stoplight System Video

The key point is that the stoplight should not stand alone. It should be paired with clear written expectations.

Pause + Do
Think about one assignment, discussion post, reflection, or professional task from your own teaching. Decide whether it is best categorized as Red light, Yellow light, or Green light. Then write 2 to 3 sentences explaining why.
Suggested Google Classroom response: Identify the task, assign it a light, and briefly explain your reasoning based on the actual learning goal.
What Faculty Need to Clarify
If we want students to use AI responsibly, we must decide what responsible use looks like.
Whether AI use is allowed, limited, or not allowed
What parts of the process AI can support
What parts need to remain fully human-authored
Whether students need to disclose or reflect on their use
Why those boundaries exist in relation to the purpose of the task
This is where tools and decision systems connect directly to teaching practice. Clearer expectations can reduce confusion, inconsistency, and avoidable problems. For many of us, this may also mean revisiting our D2L directions, assignment sheets, or syllabus language.
One practical example from teacher preparation: A candidate might use AI to help organize dictated notes about planning and preparation or professional growth into a coherent summary. In that case, I would likely see that as a yellow-light use because the goal is reflection and documentation, not polished essay writing. The candidate would still need to verify accuracy, revise the language, and make sure the final submission reflects their actual practice.
Prompt Frameworks for Task Design
A lot of weak AI output starts with weak prompting. One simple framework that works well for faculty is:
Task
What do you want the AI to do?
Context
What course, discipline, audience, or teaching situation should shape the response?
Constraints
What boundaries matter here, such as length, tone, grade level, standards, accessibility, or what should be left out?
Criteria
What would make the result actually useful?
Iterate
What needs to be revised, improved, or corrected after the first output?

Good prompting is really about clearer task design. In that sense, prompting and instructional planning are not all that different.

MIT Sloan Teaching & Learning Technologies

Effective Prompts for AI: The Essentials - MIT Sloan Teaching & Learning Technologies

Unlock the potential of AI by crafting effective prompts. Learn how prompt engineering can optimize your AI interactions, enhance output quality, and understand its limitations.

Medium

9 Best Prompting Frameworks to Supercharge Your Everyday Research with LLMs

Unlock Efficiency and Accuracy with These AI Prompting Strategies

Pause + Do
Take one prompt you have used before, or one you could realistically use in your role, and revise it using: Task, Context, Constraints, Criteria, and Iterate.
Suggested Google Classroom response: Post your original prompt and your revised version. Then add 2 to 3 sentences explaining what improved when you added more specificity.
How to Use Multiple AI Tools to Update A Presentation
From Vague Prompt to Useful Prompt
Vague Prompt
Create a lesson on rhetorical appeals.
Stronger Prompt
Create a 20-minute 9th-grade ELA mini-lesson on rhetorical appeals for a mixed-readiness class. Include an anticipatory set, one modeled example, guided practice, one formative assessment, and a short exit ticket. Keep the language accessible and align the task to critical analysis rather than memorization.
What changed?
Clearer Task
Clearer Context
Clearer Constraints
Clearer Instructional Purpose

The difference is not just length. It is instructional specificity.
Try the University of Arizona AI Platform
Now that we have looked at tools, decision systems, and prompting, the next step is to try the University of Arizona AI environment itself.
Go to chat.ai2s.org and spend a few minutes testing the tool with one real faculty-facing task. You might try one of the following:
Draft a professional email
Draft a professional email you need to send.
Organize meeting notes
Organize notes from a meeting into action items.
Generate a class outline
Generate a rough outline for a class session.
Revise writing for clarity
Revise a block of writing for clarity and tone.
Brainstorm discussion questions
Brainstorm discussion questions for a reading.
Draft assignment directions
Create a first draft of assignment directions or turn a rough idea into a more structured prompt.
As you test it, pay attention to:
1
Which model you used
2
What kind of prompt you gave it
3
What it did well
4
Where it fell short
5
Whether it actually saved you time

Pause + Do
Try out various prompts with each tool of your choosing on the University AI Platform.
Suggested Google Classroom response: What did you try in the U of A AI tool, which model did you use, and what was your honest impression of the result?

Closing takeaway: The goal is not to use AI for everything. The goal is to make better decisions about when it is useful, when it is not, and how to communicate those expectations clearly. Start with campus-supported tools when possible, match the tool to the task, and keep human judgment in the loop.
Video Library/Resources
Magicschool Videos:
Canva Code for Me Videos

End of Module
You have completed the AI Tools and Decision Systems module of the University of Arizona Educator/Instructor Preparation Programs AI Learning Series.
For now, this is the last module in the series. Check back for more updates soon.
Credits: Created with images by brent coulter — "Sonoran Sunset" • Jayeda akter — "HUMAN-IN-THE-LOOP isolated on Transparent Background"