How Teachers Detect AI in GCSE PE Written Papers

When teachers talk about AI detection in GCSE coursework, they usually mean English, History, Geography. The written component of GCSE Physical Education rarely comes up in those conversations - and that gap is worth thinking about. PE's written papers cover anatomy and physiology, health and fitness principles, applied sport science, and in some specifications sport psychology. These are topics with a large volume of clear, well-structured AI-friendly text available online. For a student under pressure, generating a 400-word answer on cardiac output or the impact of arousal on performance is exactly the kind of task that feels low-risk to hand to an AI. The process for checking PE written work should be the same careful, fair one you would use in any other coursework subject; what changes are the questions you ask.

Why PE Written Work Is a Target

The written component of GCSE PE sits in an odd position for many students. They chose PE because they are physically capable and motivated; the written exam feels like an obstacle rather than an expression of what the subject is about. For students who struggle with extended writing, the combination of unfamiliar technical vocabulary - synovial joints, VO2 max, somatotypes, the FITT principle - and the pressure to produce coherent analytical answers under time constraints makes the written component the part they are most likely to seek help with.

AI tools handle PE written content particularly well. The factual content of GCSE PE specifications is consistent, well-documented, and widely covered online, which means AI-generated answers to typical PE questions often read as fluent, accurate, and well-structured. That fluency is precisely what makes them difficult to distinguish from a capable student's genuine work - and precisely why a detection tool provides useful evidence where your own read might not flag anything immediately.

The coursework risk is somewhat lower in PE than in essay-heavy subjects because much of the written assessment happens under controlled conditions in an exam hall. But homework tasks, revision notes submitted for feedback, controlled assessment written elements, and personal performance analysis reports are all surfaces where AI use is plausible and worth checking on.

What GradeOrbit's Likelihood Score Tells You

GradeOrbit's AI detection, on school accounts, returns a score from 0 to 100 on every piece of work you check, with a Low, Medium or High confidence label, the signals it noticed and a plain-English reasoning paragraph. The score measures how far the new work departs from that student's own earlier writing on file: low means consistent with it, high means it differs in ways worth a closer look. It does not tell you with certainty whether a specific student used AI; it tells you where the balance of evidence sits, and how unusual the writing is relative to what this student has produced before.

For GCSE PE written work, two things are worth keeping in mind when reading a score. First, well-written, technically accurate answers naturally share some features with AI output - a student who has genuinely revised and absorbed the specification's technical vocabulary may produce writing that the model finds harder to distinguish from generated text. Second, PE written responses are often shorter than English or History essays, which means there is less text for the model to work with. Both factors counsel against treating a single score as a verdict. The score is a prompt to look more closely, not a conclusion.

On a GradeOrbit school account, unlimited AI detection is included in every teacher seat, whether it's lower-stakes homework or a formal coursework element heading into centre-assessed grade moderation, with no per-check charge. Rather than a generic pass/fail-style score, under your school's signed Data Processing Agreement (DPA) it compares the writing against that student's own earlier work on file - a more useful signal for exactly the well-revised, technically accurate answers this subject produces.

Reading a High Score in Context

A score above 70% on a GCSE PE written piece warrants a closer look. The first thing to do is compare it with work you know is the student's own - a previous homework, a classwork response, a paragraph written under your supervision. Read the two pieces side by side. Does the flagged work sound like the student? Does the sentence structure and vocabulary match their usual level? Does it contain specific knowledge, personal observations, or examples that you would associate with this student's experience of the subject - a sport they play, a topic you covered with particular emphasis?

A student who has genuinely absorbed the content of a PE unit tends to write about it in a way that reflects their own experience of learning it. They make the same errors their peers make, they emphasise the parts of the topic that stuck in lessons, and they occasionally confuse terms in characteristic ways. AI-generated text tends to be more evenly competent - well-structured throughout, technically accurate, but without the unevenness that reflects a real student working through material they have partly understood and partly remembered.

If the flagged piece is notably more fluent and technically precise than anything else the student has produced, and the score is high, that combination is worth investigating. If the piece is consistent with the student's usual level and the score is elevated, the score is likely reflecting the student's genuine technical accuracy rather than AI generation. As covered in our guide on how AI detection likelihood scores work, the number is one input - the rest of the evidence picture is built from what you know about the student and the work.

Starting a Fair Conversation With the Student

If the score and your own read of the work both raise concerns, the next step is a conversation - not an accusation. The approach that serves you best is the same regardless of subject: open with the work itself, not the detection result. Ask the student to talk you through a specific section. "Can you explain what you meant here by the relationship between heart rate and stroke volume?" or "Walk me through how you came up with this analysis of the athlete's performance" are questions that a student who wrote the work can answer, and that a student who did not will find very hard.

Keep the conversation low-key and specific. You are not conducting a formal hearing; you are doing what any teacher would do when a piece of work surprises them - asking the student to demonstrate their understanding. A student who has used AI will often struggle to explain the content in their own words, or give an explanation that does not match the sophistication of what they submitted. A student who wrote their own work will usually talk about it with the confidence and imprecision of someone who knows the topic imperfectly - which is exactly right for a GCSE student.

The full framework for handling a high score from the initial detection through to the student conversation and beyond is covered in how teachers handle a high AI likelihood score fairly. The PE context does not change the core process; it only changes the subject-specific questions you might ask to test understanding.

Documenting Your Decision for Moderation

Whatever outcome your conversation produces, write it down at the time. A contemporaneous record is far more useful than a reconstruction from memory if the case is ever revisited during moderation or a student challenge. The record does not need to be long - a few sentences noting the score, the nature of your concern, what you asked the student, what they said, and what decision you reached is sufficient.

If you are satisfied that the work is the student's own, note that explicitly: "Likelihood score 73%; consistent with prior work on this topic; student explained content fluently in conversation; no further action." If you remain concerned, note that too, and follow your school's academic integrity policy for the next step - whether that is escalation to the head of department, referral to the exams officer, or a formal re-sit under supervised conditions.

For work heading into any form of moderation, include the detection run alongside your marking notes. A moderator reviewing borderline PE coursework benefits from knowing that you applied due diligence, and detection results fit into the broader evidence pack for centre-assessed and internally marked components alongside that.

Your school's academic integrity policy is the framework within which all of this sits. If PE is not explicitly mentioned, apply the same process the policy describes for coursework or extended writing - the medium is different but the principles are identical. If your school does not yet have a policy that covers AI detection, the school AI academic integrity policy guide is a useful starting point.

Running PE Written Work Through GradeOrbit

On a school account, you can check one pupil from their profile or a whole class at once. For a class set, bulk upload splits the pages by pages per student, and GradeOrbit reads the name on each cover page and matches it to a pupil in that class; anything it cannot match confidently is left for you to assign. Each answer is then compared with that pupil's own earlier work on file rather than with a generic sample, and a pupil with no earlier work on file is skipped, with you told.

For handwritten exam-style responses, scan or photograph the pages clearly and upload the images directly, or use the phone QR camera link. For typed or word-processed submissions, export to PDF and upload that (Word files aren't accepted). The AI reads the page images directly, so the format of the work does not matter. Results appear on each pupil's profile with a score, a confidence label and a plain-English summary of the signals it noticed. You retain all control over what happens next - the tool provides the score, and you exercise your professional judgment about what it means in the context of this student and this piece of work.

Because AI detection compares written work against earlier work on file, it runs on GradeOrbit school accounts under a signed Data Processing Agreement (DPA), where pupils' work is held with their real names: uploaded scans are removed when a run archives (after 90 days by default, adjustable from 7 to 365 days), pupil records follow your school's retention window, and everything is wiped at offboarding. Individual teacher accounts do not include detection; they cover marking, where uploaded student work is redacted client-side in your browser, labelled anonymously, and never stored (processed and discarded immediately). Across every tier, student work is never used to train AI models.

Summary for PE Departments

PE written work is fluent, factual and easy for an AI to imitate, so a high score needs more context here than in most subjects: compare against the student's own earlier work, ask them to explain a specific answer, and write down what you found. Detection on GradeOrbit for Schools works the same way for PE as for any other subject, and school accounts can start with a free school trial so a department can try it on real work first.