AI Tools for PE Teachers: Honest Review After Testing 5 Tools for 6 Weeks

I want to be honest about what happened. I walked into the gymnasium with 24 eighth graders, a whistle I didn't know how to use effectively, and a sticky note that said "fitness unit — they know what to do." They did not know what to do. What followed was 47 minutes of managed chaos that I am not proud of and will not describe in detail.

What I will say is this: somewhere around minute 23, when I was trying to explain a circuit training rotation to students who were already throwing dodgeballs at each other, I thought — PE teachers do not get nearly enough credit for what they manage every single day.

That experience sent me down a rabbit hole. PE teaching has a specific set of challenges that no other subject shares — you're managing physical safety, student fitness data, movement assessments, health education content, and lesson planning for a gymnasium full of bodies in motion, all simultaneously. I wanted to know whether AI tools for PE teachers actually addressed any of those specific challenges or whether they were just generic education tools with a sports graphic on the landing page.

Six weeks. Five tools. Here's what I found.

Why PE Teaching Needs Its Own AI Conversation

Physical education is the most consistently under-resourced and professionally underestimated subject in most school buildings. According to the Society of Health and Physical Educators (SHAPE America), the recommended standard is 150 minutes of PE per week for elementary students and 225 minutes for middle and high school students. A 2023 CDC report on school health policies found that fewer than 25% of U.S. states require those recommended minutes — meaning most PE teachers are fighting for instructional time while simultaneously being expected to document student fitness progress, teach health literacy, differentiate for students with physical disabilities, and manage the administrative load of any other teacher.

That administrative load is where AI tools have the most immediate potential. Lesson planning for a gymnasium is genuinely different from lesson planning for a classroom — you're working with space constraints, equipment availability, weather (for outdoor units), student fitness levels that vary enormously, and safety considerations that don't exist in a math room. A good AI tool for PE teachers understands those constraints. Most don't. The ones that do are worth knowing about.

My Testing Methodology

Testing period: March 10 – April 18, 2025.

I tested five AI tools across four PE-specific use cases:

• Lesson planning for gymnasium and outdoor settings
• Fitness assessment tracking and data interpretation
• Health education content creation (the classroom-based component of PE)
• Differentiation for students with physical disabilities or modified participation needs

I worked alongside two PE colleagues — one with 11 years of experience teaching middle school PE, one in her third year teaching high school health and PE — who tested tools in their actual classes and gave me feedback on real-world usability.

Tools tested: ChatGPT with PE-specific prompts, MagicSchool AI, Plankk Studio, Google's NotebookLM, and Canva for health education content. All tested on free or trial tiers. Paid features noted where relevant.

Evaluation criteria: relevance to PE-specific challenges, output quality for intended use, time saved versus manual approach, and usability during or immediately before class.

Data privacy note: Student fitness data is protected under FERPA and, in many states, additional health privacy regulations. No student fitness records or health information were shared with any AI platform during testing. PE teachers should consult their district's data privacy officer before using any AI tool that involves student health or fitness data.

What Actually Worked

1. ChatGPT With PE-Specific Prompts — Best for Lesson Planning

This is the finding that will surprise most PE teachers: the general-purpose tool, prompted correctly, outperformed every PE-specific tool I tested for lesson planning. Not because PE-specific tools are bad — one of them is genuinely useful — but because lesson planning for PE requires so many specific constraints that a conversational AI handles them better than a form-based generator.

Here's the problem with form-based PE lesson generators: they ask for grade level, activity type, and duration. That's it. What they don't ask — and what actually determines whether a PE lesson works — is equipment available, gymnasium dimensions, number of stations, maximum students per station for safety, fitness level range in the class, any students with modified participation plans, and the specific SHAPE America standard being addressed.

ChatGPT, given all of that context in a prompt, produces lesson plans that account for every constraint simultaneously.

The prompt structure I developed with my PE colleagues:

"Write a PE lesson plan for [grade level] students. Topic: [fitness unit/sport/movement skill]. Duration: 50 minutes. Equipment available: [specific list]. Space: gymnasium with four basketball courts, no divider curtains. Class size: 28 students. Fitness level range: mixed — approximately one-third high fitness, one-third moderate, one-third low. Three students have modified participation plans: one uses a wheelchair, one has a knee injury limiting running, one has an asthma condition requiring modified intensity. Address SHAPE America Standard [number]. Include: warm-up (8 minutes), skill instruction (10 minutes), practice rotations (25 minutes), cool-down and reflection (7 minutes). Note specific modifications for the three students with modified participation needs at each stage."

The output from this prompt for a cardiovascular fitness lesson was detailed enough that my 11-year PE colleague — who reviewed it without knowing the source — said it was "better structured than what most student teachers submit for their practicum placement." Her words exactly.

The modification section was the strongest part. For the wheelchair user, the prompt produced specific upper-body cardiovascular alternatives at each rotation station rather than generic "modify as needed" notes. That specificity matters in a gymnasium where a teacher is managing 28 students simultaneously and cannot stop to improvise modifications on the spot.

Output quality: 9/10 with specific prompt PE-relevance: 9/10 when constraints are fully specified Time to usable lesson plan: 12–18 minutes including prompt construction Free tier: Yes

2. MagicSchool AI — Best for Health Education Content

PE teachers teach two distinct things: physical activity in the gymnasium and health education in the classroom. The classroom component — nutrition, mental health, human development, substance prevention, personal safety — is where MagicSchool AI earns its place in a PE teacher's toolkit.

My third-year colleague was preparing a mental health unit for her 10th grade health class. She had the content knowledge but was spending significant time finding age-appropriate scenarios for discussion activities, writing quiz questions that didn't feel clinical or preachy, and building a rubric for a personal wellness reflection assignment. All three of those tasks were taking her evenings.

I sat with her for one planning session using MagicSchool AI. Here's what we generated:

Scenario-based discussion prompts for mental health unit: MagicSchool generated eight realistic teen scenarios depicting different mental health situations — stress, social anxiety, grief, peer pressure — written at a reading level appropriate for 10th graders without being either simplistic or clinically loaded. We used six of them directly. Time: 7 minutes.

Health quiz questions: MagicSchool generated a 12-question mixed-format quiz on stress management and coping strategies with Bloom's Taxonomy levels specified. We used ten questions with minor edits. Time: 9 minutes.

Wellness reflection rubric: Using the qualitative-descriptor prompt structure I've documented in previous reviews, MagicSchool generated a four-category rubric for a personal wellness reflection that assessed self-awareness, goal specificity, evidence use, and written communication. The descriptors were qualitatively distinct across levels. We used it without edits. Time: 11 minutes.

Total time for three tasks that would have taken my colleague a full evening: 27 minutes. She used all three outputs in class the following week.

Output quality for health education content: 9/10 PE gymnasium relevance: Limited — use for health education component only Time saved: 60–90 minutes on health education content creation Free tier: Yes, with daily usage limits

3. Canva — Best for Health Education Visual Materials

PE and health teachers create more visual content than almost any other subject area — workout posters, nutrition charts, heart rate zone diagrams, safety instruction cards for equipment, wellness infographics for bulletin boards. Most of this gets made in PowerPoint at 9pm and printed on a school printer with a toner cartridge that's been "low" since October.

Canva's AI-assisted design tools — Magic Design, Magic Write, and the template library — don't generate PE content from scratch the way MagicSchool generates lesson plans. What they do is make professional-quality visual materials accessible to teachers who don't have graphic design training and don't have time to learn it.

Specific applications I tested and used in real settings:

Gymnasium station instruction cards: I described each station in a circuit training rotation to Canva's Magic Write, specified the format (bold exercise name, numbered steps, modification note, target muscle group), and generated clean card text. Then I applied a Canva sports template with clear visual hierarchy. Printed and laminated: 8 station cards that students could follow independently without the teacher explaining each one. Setup time: 22 minutes for all eight cards. My PE colleague estimated she previously spent 45–60 minutes making similar cards in PowerPoint.

Nutrition education infographic: For a health education lesson on macronutrients, Canva's template library had three appropriate infographic formats. Magic Write generated the content text based on a brief I provided. The finished infographic — appropriate for classroom display and student handouts — took 18 minutes. The previous version my colleague was using was a photocopied sheet from a textbook published in 2014.

Heart rate zone poster: A visually clear heart rate zone chart for the gymnasium wall took 14 minutes. Students referenced it during cardio rotations for the remainder of the unit.

Output quality: 8/10 for visual design, 7/10 for AI-generated content (always review health facts for accuracy) PE-relevance: High for visual materials — limited for lesson planning Time saved: 30–45 minutes per visual material set Free tier: Yes — Canva's free tier is sufficient for most PE visual needs

4. NotebookLM — Best for PE Teacher Professional Development and Research

This application surprised me. NotebookLM's strength — synthesizing uploaded documents and answering questions from them — has a specific PE use case that isn't immediately obvious: helping PE teachers navigate the research and standards documents that govern their practice.

SHAPE America's National Standards and Grade-Level Outcomes document is 230 pages. The PE curriculum frameworks published by most state education departments add another 80–150 pages. Most PE teachers have read portions of these documents but don't have them internalized at a level that makes standards alignment in lesson planning fast and accurate.

I uploaded the SHAPE America standards document and my state's PE curriculum framework into a NotebookLM notebook. Then I asked it questions that a PE teacher would actually ask during lesson planning:

"What does Standard 3 require students to demonstrate by the end of 8th grade, and what evidence of learning would satisfy that requirement?"

"My lesson involves cardiovascular fitness stations. Which SHAPE America standards does this address and what are the grade-level outcomes for 7th grade?"

"Generate three learning objectives for a badminton unit at the 9th grade level that align to SHAPE America Standards 1 and 2."

Every answer was sourced directly from the uploaded documents with page citations. My PE colleague with 11 years of experience said: "I've been teaching to these standards for a decade and I still would have had to look three things up. This answered all of them in 30 seconds."

That's professional knowledge acceleration. Not replacing expertise — accelerating access to the standards documentation that expertise is supposed to be built on.

Output quality: 9/10 for standards navigation PE-relevance: High for professional development, moderate for direct lesson planning Time saved: Significant on standards research and alignment tasks Free tier: Yes

What Didn't Work

Plankk Studio — Wrong Problem, Right Idea

Plankk Studio is a fitness platform that uses AI to generate workout programs. I included it because several PE teacher communities online recommend it and I wanted to test whether a fitness AI could cross over into educational settings.

It can't. Not meaningfully.

Plankk generates workout programs optimized for individual fitness goals — the kind of thing a personal trainer uses with a client. It does not understand the educational context of PE: the need for SHAPE America alignment, the requirement to differentiate for a class of 28 students simultaneously, the safety and supervision constraints of a gymnasium, the curriculum progression across a unit, or the assessment documentation requirements of school-based PE.

The workout programs it generated were physically sound. They were educationally irrelevant. A PE teacher would need to completely reconstruct the output as a lesson plan — which means Plankk saved no time and added a translation step that a good ChatGPT prompt eliminates entirely.

This is the fitness app problem with PE AI tools generally: the fitness industry and the education industry have different goals, different constraints, and different outputs. A tool built for one doesn't transfer to the other without significant redesign.

The Moment That Reframed My Thinking

Four weeks into testing, my 11-year PE colleague was reviewing a ChatGPT-generated lesson plan for a pickleball unit. The plan was structurally strong — warm-up, skill instruction, practice progression, cool-down, assessment checkpoint. She read it carefully, nodded at several points, then stopped at the practice progression section.

"This has students practicing the backhand drive before they've established a consistent ready position. That's backwards. Nobody who actually teaches beginners would sequence it this way."

She was right. ChatGPT had produced a technically complete lesson plan with a pedagogically incorrect skill progression — because skill progression in sport pedagogy follows developmental logic that requires domain expertise to verify. The AI knew what pickleball skills existed. It didn't know the correct order to teach them to beginners.

This is the PE-specific version of a problem that exists in every subject: AI tools generate plausible sequences, not necessarily correct ones. In a math lesson, a wrong sequence is a teachable moment. In a PE lesson, a wrong skill progression builds compensatory movement patterns that are harder to correct later and, at the extreme end, increase injury risk.

Every AI-generated PE lesson plan needs review by someone with physical education pedagogy knowledge before it goes into a gymnasium. This is non-negotiable. The tool drafts. The trained PE teacher verifies.

The PE Lesson Plan Review Checklist

Based on six weeks of testing and my colleagues' expertise, here's the review checklist I recommend for every AI-generated PE lesson plan before it reaches students:

Skill progression check: Is the sequence of skill instruction developmentally appropriate — simpler skills before complex ones, static practice before dynamic application?

Safety scan: Does the lesson account for the specific safety requirements of the activity — spacing, equipment handling, student density per station?

Modification completeness: Are modifications for students with physical disabilities or modified participation plans specific enough to implement without improvisation during class?

SHAPE America alignment: Does the stated standard actually match the activity — not just the general topic?

Time realism: Are the time allocations realistic for actual gymnasium transitions — equipment distribution, formation changes, and instruction time with a live group of students?

Assessment checkpoint: Is there a formative assessment moment built in that generates actual data about student skill development — not just participation?

Six checks. Ten minutes. Every AI-generated PE lesson plan. Every time.

My Actual PE AI Workflow (As Recommended to My Colleagues)

Gymnasium lesson planning: ChatGPT with the full constraint-specific prompt. SHAPE America standard specified. Modifications named individually. Review checklist applied before use. 15–20 minutes per lesson plan.

Health education content: MagicSchool AI for scenarios, quiz questions, and rubrics. 7–11 minutes per content type.

Visual materials: Canva for station cards, posters, infographics, and handouts. 15–25 minutes per visual set. Always verify health facts for accuracy.

Standards navigation and professional development: NotebookLM with SHAPE America and state curriculum documents uploaded. Query during planning for fast standards alignment. Zero additional time once notebook is set up.

Total weekly time saved across all four use cases: approximately 3–4 hours for a full-time PE and health teacher managing multiple classes and units simultaneously.

Who Benefits Most From AI Tools for PE Teachers

PE teachers managing both gymnasium and health education components will see the widest return — MagicSchool and Canva for health education content, ChatGPT for gymnasium lesson planning, NotebookLM for standards navigation. The tools address different parts of the job and work best in combination.

New PE teachers still building their pedagogical content knowledge should treat AI-generated lesson plans as drafts that require mentor review — not finished products. The skill progression problem my colleague identified is the kind of error that an experienced PE teacher catches immediately and a first-year teacher might not. Use the tools to save time on structure and formatting; rely on mentorship for pedagogical sequencing verification.

PE department heads building unit-level curriculum maps: NotebookLM with your state's curriculum framework uploaded can dramatically accelerate standards alignment documentation — the part of curriculum work that takes the most time and produces the least direct instructional value.

Final Verdict

AI tools for PE teachers are genuinely useful — but only when they're used with an understanding of what PE teaching actually requires. The best tool for gymnasium lesson planning isn't a PE-specific app. It's a general AI given enough specific context to understand the constraints of your gymnasium, your students, and your standards. ChatGPT with a well-constructed prompt does this better than any purpose-built PE tool I tested.

MagicSchool AI and Canva handle the health education component cleanly and save real time. NotebookLM is quietly one of the most useful professional development tools available for any teacher navigating large standards documents — PE teachers especially.

And Plankk Studio, for all its fitness sophistication, reminded me of something worth saying plainly: a tool built for the fitness industry is not a tool built for education. The goals are different. The constraints are different. The outputs need to be different. Not every AI tool belongs in a classroom just because it involves movement or health.

I still don't know how to use a whistle effectively. But I know considerably more about what AI tools can and can't do for the teachers who can.