🤖 A teacher AI prompting framework

🤖 Get the teacher AI literacy essentials

FYI: The Teacher AI Literacy Level 1 course is available for pre-registration right now.

• It’s 50% off through July 7.

• On July 7, the content becomes available … and you can keep it forever.

• It’s hosted by me, Ken Shelton, and Holly Clark — three AI edu authors and speakers who have trained a combined 1 million educators.

• It’s asynchronous with a certificate of completion (likely 8-10 hours of PD credit).

Looking for guidance on how AI impacts education, schools, and the classroom — and how to navigate it responsibly? This course is for you.

📚 New AI resources this week

1️⃣ We're professors. We're parents. HISD students don't deserve AI slop. (via Houston Chronicle) — Educators at Rice University and parents of HISD students are concerned about the district’s direction since its takeover by the Texas Education Agency in 2023.

2️⃣ K–12 Experts Share How to Best Embrace AI in Schools (via EdTech Magazine) — As artificial intelligence tools become ubiquitous, districts must make critical decisions that could reshape teaching and learning for years to come.

3️⃣ Exploring K-12 Physical Education Teachers' Perspectives on Opportunities and Challenges of AI Integration (via arXiv) — Participants envisioned AI playing multidimensional roles, such as an operational assistant, personal trainer, group coach, and evaluator, as solutions to address unique instructional and operational challenges in K-12 PE classes.

📢 Your voice: Keys to AI in edu

Last week’s poll: Which of these keys is most important for AI in edu?

🟨🟨🟨🟨🟨⬜️ Human thinking before AI (17)
🟨⬜️⬜️⬜️⬜️⬜️ Keep the main thing the main thing (4)
🟨🟨⬜️⬜️⬜️⬜️ Always be critiquing (8)
🟩🟩🟩🟩🟩🟩 Focus on process over product (18)
⬜️⬜️⬜️⬜️⬜️⬜️ Other ... (0)

Process over product: We need to show students/teachers that AI will produce a “thing” at the end, and it’s often mediocre at best. We need to really help them understand the PROCESS of taking an initial output and making it something better and more useful. — Cammie Kannekens

Process over product: I feel like we get so focused on the end result or product that we lose sight of what it takes to get there; the learning and the journey in the process can and should be more valuable. The struggle, perseverance, thinking, collaborating, and all of the layers involved from beginning to end can be equally important. We need to value human thinking before AI, no doubt - I also think there's value in using a resource available to us to supplement our thinking, not replacing it. — Chad Sussex

Always be critiquing: In the information age, there always needs to be humans critiquing. My wife blames mistakes she sees in publications on AI. I try to explain that it's not a failure of the AI but of the human who didn't verify the information. There seems to no longer be editors. She showed me an article about Billy Crystal accompanying Michael J. Fox to a basketball game. In the picture, it was clearly Bill Murray and not Billy Crystal, yet the article kept referring to Crystal and his body of work. I explained that there was no human editor who proofed this or reviewed it. Keep the human in the loop, and the place for the human is between what the AI spits out and what is released to others. — H. Coleman

Human thinking before AI: AI is a helper, an assistant that builds ON your thinking, but isn't a replacement. This is true for the adults but also for the students--in the end, it's just a machine and it needs to be tempered with critical thinking. — J. Stuivenberg

Process over product: Prioritizing the learning process cultivates critical thinking and showcases genuine understanding. Documenting this process ensures the final product is authentic and equips students to effectively critique AI collaborations/assistance. — Virma Santiago

What would you like to read in AI for Admins?

What’s a topic you’d like to see covered here? Hit REPLY to this email and let me know.

Have you done anything you’d like to share with the AI for Admins community? Hit REPLY and let me know.

Would you like to write a guest post to support and equip AI for Admins readers? Hit REPLY and let me know.

🗳 Poll: Best practices for prompting AI

Instructions:

1. Please vote on this week’s poll. It just takes a click!

2. Optional: Explain your vote / provide context / add details in a comment afterward.

3. Optional: Include your name in your comment so I can credit you if I use your response. (I’ll try to pull names from email addresses. If you don’t want me to do that, please say so.)

👨‍💻 A teacher AI prompting framework

Originally posted in the Ditch That Textbook newsletter

I’ve been working with fellow presenters/authors Holly Clark and Ken Shelton on our new Teacher AI Literacy Level 1 course.

We have a whole section on AI prompting, and we’re developing a sort of periodic table of AI prompting.

Caution: Don’t worry about your eyesight … the image below is blurry on purpose. (Sorry … a necessary step though!)

The more that I analyze my own prompting, the more I realize that it has lots of elements … lots of pieces to the puzzle that I request from an AI assistant like ChatGPT.

Eventually, I started to see them chaining together, kind of how the elements on the periodic table bond together. (Don’t look too closely at this analogy … remember, I’m a Spanish teacher and not a science teacher!)

For the Teacher AI Literacy Level 1 course, I’m developing a sort of periodic table of AI prompting elements.

The purpose: to help us identify parts to add to our prompts so they’re richer — and so they’re more likely to get us exactly what we’re looking for.

Below are three examples of element chains — and a description of how they work …

1. Teaching Material Prompt

In this prompt, we’ll create a teaching resource — a reading passage about photosynthesis — but we’ll add a couple extra elements to level up the response.

Full prompt: You are a sixth grade science teacher. Write a short reading passage about photosynthesis at three reading levels: 3rd, 6th, and 9th grade. Return each version in simple, short paragraphs.

1. R (Role): “You are a sixth grade science teacher.” This sets the role the AI will play, a common first step in prompting.

2. T (Task): “Write a short reading passage about photosynthesis.” Describes the task in a clear and instructional way.

3. Lv (Levels): “At three reading levels: fourth, sixth, and eighth.” Creates materials at different reading levels, incorporating differentiated instruction.

4. Fm (Format): “Return each version in simple, short paragraphs.” Keeps the output clean, clear, and ready to use.

Why this prompt works: It has solid core features (role and task) that most any teacher AI prompt will need. But it adds a layer of instructional design — leveled texts — AND it requests a specific format so it requires little extra formatting afterward.

2. Classroom Instruction Prompt

In this prompt, we’ll be creating a multifaceted lesson plan that covers some key parts of instruction to help students succeed.

Full prompt: You are an eighth grade math teacher. Explain the concept of slope to students who are just being introduced to it. Include at least two real-world examples of slope, a common misconception students have about slope, and a reflection question to help them connect the idea to their lives.

1. R (Role): “You are an eighth grade math teacher.” Again, sets the role so the AI knows how to act.

2. T (Task): “Explain the concept of slope …” Defines our objective AND adds a little extra context to make sure the response is precise.

3. Ex (Examples): “Include at least two real-world examples …” In this case, we’re not just asking for example problems. We’re asking for places where you’d find this outside of school.

4. M (Misconceptions): “Include a common misconception students have …” If we know that students struggle with this new concept, let’s address it right away.

5. Rf (Reflection): “… and a reflection question …” Add a metacognitive question at the end to solidify learning.

Why this prompt works: It builds on the basics. It includes the core — role and task — but it also helps us pull in a few things we might not think about — real-world examples, misconceptions, and reflection.

3. Differentiated Questions Prompt

In this prompt, we’ll create some student questions using a bit of differentiated instruction. Plus, we’ll introduce the idea of an optional element that might fit in certain circumstances.

Full prompt: You are a fifth grade social studies teacher. Write three comprehension questions about the causes of the American Revolution. Make each question target a different level of Bloom’s Taxonomy: recall, understanding, and analysis. One question should be scaffolded for an English language learner.

1. R (Role): “You are a fifth grade social studies teacher.” As usual, establish the role right away.

2. T (Task): “Write three comprehension questions …” Simple, clear, focused task.

3. Lv (Levels): “Make each question target a different level …” Instead of reading levels, now we’re going to draw on a different model (below). But it still calls on the idea of levels.

4. Cd (Cognitive Demand): … of Bloom’s Taxonomy: recall, understanding, and analysis.” This helps students on a range of complexity of thought.

5. N (Need of Student) (optional): “One question should be scaffolded …” This factors in the needs of particular student populations or particular student needs.

Why this prompt works: It goes beyond the simple “give me 10 questions” prompt that many beginner prompters would use. It draws in important pedagogical techniques — levels, cognitive demand, and needs of student — even if the teacher doesn’t feel like an expert in those areas.

More prompts — and more AI prompting elements

We’re developing this periodic table of AI prompting elements for our Teacher AI Literacy Level 1 course in The AI Fluency Lab.

Until July 7, the course (regularly $99 USD) is on sale for 50 percent off.

You can register today to lock in the discount — and on July 7, when the materials become available, you can get started!