AI has arrived in the PE classroom. A practical guide for teachers and coaches — including what to do with no budget, what these tools are genuinely bad at, and the student-data question nobody raises.
Artificial intelligence has arrived in the PE classroom, and for physical education teachers and coaches in India this is a present reality rather than a development to watch. It is reshaping how lessons are planned, how progress is assessed and — under NEP 2020, which made PE a credit-bearing subject — how that progress is defended to an academic council.
Most writing on this subject assumes a budget for wearables and video analysis software. Most Indian schools do not have one. So this guide starts where the majority of teachers actually are.
Start here if your school has no budget
You can do the useful half of this with a phone and a spreadsheet. Nothing below requires a purchase.
- Record one skill, once a term, from a fixed position — same spot, same angle, same distance. Consistency of capture matters more than camera quality, and it is what makes any later comparison meaningful.
- Use a general-purpose AI assistant to turn a term’s scribbled observations into a structured progress summary per student. This is the single biggest time saving available to a PE teacher today, and it costs nothing.
- Ask an AI assistant to draft differentiated lesson variants — the same session adapted for a student returning from injury, a beginner and an advanced player. Review and correct them; the draft is the time saved, not the final plan.
- Build one shared sheet per class with a handful of measures you can actually collect every term. Consistent simple data beats sophisticated data collected twice and abandoned.
The schools that get value from AI later are the ones with two years of consistently recorded, boring data. The tools change every year; the record is what compounds.
The tools, grouped by what they actually do
- Video analysis — auto-tagging events in lesson footage and comparing a student’s technique against their own earlier attempts
- Wearables — heart rate, training load and recovery, useful mainly where there is a genuine question about workload
- Planning assistants — generating and adapting session plans, progressions and differentiated variants
- Dashboards — turning a term of scattered measurements into something legible at a glance
- Explanatory tutors — chatbots that answer students’ sports science questions outside lesson time
What it changes about planning
The immediate effect is on administration rather than on teaching. A teacher who spent hours assembling progress records can get to a first draft in minutes, and see which students are improving, which are flat and which might be carrying more load than is sensible.
That time moves to coaching, encouraging and the human judgement no tool replicates. AI does not replace a teacher’s instinct — it frees the hours in which that instinct does its best work.
Assessment, and why NEP 2020 raises the stakes
Assessment in physical education has always been partly subjective, and that was tolerable while PE sat outside the credit system. It is not tolerable now. A credit-bearing subject needs assessment that can be explained to a parent, an academic council and an external examiner.
This is where AI earns its place. Video-based tools can evaluate running technique, jumping mechanics or throwing form with more consistency across thirty students than a teacher watching thirty students in sequence — not because the tool sees better, but because it does not get tired, distracted or fonder of some students than others.
Consistency is the real contribution, and it is worth being precise about that. These tools make errors, they make them systematically rather than randomly, and they are less reliable on body types, clothing and settings unlike those they were trained on — which for equipment built elsewhere may well include your students. A grade should never rest on a tool’s output alone.
What AI is genuinely bad at in a PE setting
- Anything requiring context — it cannot tell a lazy sprint from a cautious one three days after an ankle sprain
- Group dynamics, effort and attitude, which is most of what a PE teacher is actually assessing
- Safety judgement — no tool should decide whether a student continues after a knock
- Unusual bodies and unusual techniques, where confident, wrong output is the failure mode rather than an error message
- Anything where you cannot check the answer — if you could not have reached a similar conclusion yourself with more time, you cannot supervise the tool
Student data: the part nobody discusses
Every item in this article involves recording children. Video of minors, heart-rate traces and body measurements are personal data, and India’s Digital Personal Data Protection Act sets specific obligations around processing a child’s data, including verifiable parental consent.
Before a single recording, settle four things with your school: who has consented and in writing, where the footage is stored and for how long, whether the tool uploads it to a server outside the school, and what happens to it when the student leaves. A free app that processes video in the cloud is making a decision about your students’ data on your behalf.
None of this is a reason to avoid the tools. It is a reason to have the conversation before the first lesson rather than after the first complaint.
A realistic first term
- Weeks 1-2 — pick ONE thing to improve. Progress records or a single skill assessment. Not both.
- Weeks 3-4 — settle consent and storage with your school before recording anything.
- Weeks 5-10 — collect the same simple measures every week, in the same way, without changing the method.
- Weeks 11-12 — use an AI assistant to summarise the term per student, then read every summary and correct it. The corrections are where you learn what the tool is bad at.
One class, one term, one measure. Teachers who try to instrument everything at once produce a dataset nobody trusts and abandon it by the second term.
What AI literacy actually means here
The shift AI asks of a PE professional is conceptual rather than technical. You do not need to understand how a model works. You need to know what it can and cannot do, to ask sensible questions of your own data, to recognise a confident answer that is wrong, and to keep professional judgement above the tool’s recommendation.
That is a teachable skill, and it is becoming as much a part of the job as understanding anatomy or training theory. It is also the part that stays valuable when the specific tools are replaced, which they will be.
Where this sits in a degree
Sportal Corporate’s B.P.E.S. carries AI tools for physical education as part of the curriculum rather than as a separate certificate — the same approach as the sports analytics in our B.S.M. The argument for building it in is the one above: the tools will change several times over a four-year degree, so what has to be taught is the judgement, not the software.
Explore the programme, or register your interest on WhatsApp.