How Teachers Detect AI in GCSE Design and Technology Coursework
GCSE Design and Technology is, on paper, a practical subject - sketches, prototypes, a finished product. But the non-examined assessment that carries half the qualification is built around a written design portfolio: the analysis of a context, the research into existing products, the evaluation of a client brief, the justification of design decisions, and the testing and review of the outcome. That written commentary is where the marks sit, and it is also the easiest part of the whole NEA for AI to quietly produce. If you teach D&T on the AQA, Edexcel, OCR, or Eduqas specifications, a fair detection check on that writing is part of protecting the portfolio - provided the score is never treated as a verdict.
Why the D&T Portfolio Is Exposed
The making is supervised. Students cut, shape, solder, and assemble in your workshop under your eye, so the physical outcome is hard to fake. The writing is not. Context analysis, product research, design justification, and evaluation are overwhelmingly produced at home, across weeks, on the same laptop a student uses for everything else. The prompts those sections answer - "analyse the needs of your client," "evaluate two existing products against your brief," "justify your choice of material and manufacturing process" - are exactly the kind of structured, explain-and-justify writing that AI tools generate fluently and at length.
The subject content compounds it. Material properties, manufacturing processes, the work of named designers, environmental and sustainability considerations - all of this is extensively documented online, which means a generative tool can produce a confident, well-organised passage on the working properties of plywood or the principles of mass production with almost no prompting. That fluency is what makes AI-written portfolio commentary difficult to spot on a first read, especially when the practical work it sits alongside is genuinely the student's own.
None of this means most students are misusing AI. The overwhelming majority design, make, and write honestly. But the portfolio is half the GCSE, it is internally marked and externally moderated, and a routine, fair detection check protects the students who did the work properly from being graded on the same scale as those who did not.
What GradeOrbit's Likelihood Score Tells You
GradeOrbit's AI detection, which comes with its school accounts, returns a score from 0 to 100 on each piece you check. The score measures how far the portfolio writing departs from that student's own earlier written work on file: low means it is consistent with how they usually write, high means it differs in ways worth a closer look. It comes with a Low, Medium or High confidence label, the signals it noticed and a plain-English explanation. It does not tell you with certainty that a particular student used AI; it is an indicator to review, never proof.
For a D&T portfolio there are two things worth holding in mind. First, technical writing about materials and processes reads differently from a student's usual essays - formal vocabulary, list-like structure, confident factual claims - so a competent student writing a strong specification section may sit a little higher than you would expect. Second, run the analytical and evaluative sections, not the bullet-point spec tables: extended prose gives the model more to work with and produces a more informative score than a few labelled diagrams or a parts list. Both point the same way - the score is a prompt to look more closely, not a conclusion to act on.
AI detection is included with unlimited checks in GradeOrbit's School teacher seats, so there are no tiers to choose between and no per-check charge - departments can run checks across an entire cohort's portfolios and again on the final NEA heading into moderation. Because School accounts operate under a signed Data Processing Agreement (DPA), each portfolio is compared against work that student has already had marked in GradeOrbit, providing a grounded comparison rather than an isolated score. A student with no earlier work on file is skipped, and you are told.
Reading a High Score in Context
A score above 70% on a portfolio section warrants a closer look. The most useful comparison you have is the student's own work produced in front of you: the design sketches annotated in lessons, the workshop diary, the short written tasks completed in class. Read the flagged commentary alongside that supervised work. Does the analysis sound like the same student? Is the written sophistication consistent, or does the portfolio leap to a fluency and technical command that their classwork never approaches?
Genuine D&T writing is usually anchored to the specific project. Students refer to the actual client they were given, the particular constraints of the brief, the material they could get hold of in your workshop, the thing that went wrong with their first prototype. AI-generated commentary is often more evenly competent and oddly detached: technically accurate about plywood in general, but vague about this student's plywood, this joint, this test. A passage that discusses manufacturing processes the student never used, in a register their classwork never reaches, is worth investigating when the score is also high.
If the flagged section is broadly consistent with the student's supervised work and the score is merely elevated, the number is most likely reflecting a strong, genuine writer. As covered in our guide on how teachers handle a high AI likelihood score fairly, the score is one input - the rest of the picture comes from what you know about the student and the project.
Starting a Fair Conversation With the Student
If both the score and your own read raise concerns, the next step is a conversation, not an accusation. The most reliable approach is to open with the design work, not the detection result. Ask the student to talk you through a decision: "Walk me through why you chose this material over the alternative here," or "What did your testing actually show, and what did you change because of it?" A student who genuinely researched, designed, and wrote the portfolio can reconstruct the reasoning. A student who generated it usually cannot connect the written justification back to the object on the bench.
Keep the tone low-key and specific. You are doing what any D&T teacher does when a piece of written work does not match the maker in the workshop - asking the student to demonstrate the thinking behind it. Students who did the work talk about it with the confidence and the gaps you would expect from someone who wrestled with a real brief over several weeks. That texture is hard to fake and very revealing.
Documenting Your Decision for Moderation
GCSE D&T coursework is internally marked and externally moderated, so contemporaneous documentation matters. Whatever your conversation concludes, write it down at the time. A few sentences will do: the score, the nature of your concern, what you asked, what the student said, and the decision you reached. A record made on the day is far more defensible than one reconstructed from memory if the case is revisited.
If you are satisfied the work is the student's own, note that explicitly - "Detection score 76 on the research section; consistent with workshop diary and supervised tasks; student explained the material choice fluently; no further action." If concerns remain, follow your centre's academic integrity policy and the awarding body's malpractice procedures for the next step. Including the detection run in your marking notes shows a moderator that due diligence was applied; the moderation cycle guide covers how detection results sit within the wider evidence pack for internally assessed components, and if your centre's policy does not yet address AI, the school AI academic integrity policy guide is a good place to start.
Running Coursework Through GradeOrbit
On a school account, a class set of D&T portfolios can be checked in one go. Bulk upload splits the pages by the number of pages per student, then reads the name on each cover page and matches it to your class list, asking you to pick the pupil if the match isn't clear; for a single portfolio, run the check from that pupil's profile instead. Each portfolio is checked against that pupil's earlier written work on file, and the result appears on their profile. Typed portfolios go in as a PDF - export from Word or PowerPoint first, as Word files aren't accepted - and handwritten design sheets as clear scans or photos, which you can take on your phone through the QR camera link. The same workflow applies whether you are checking one final submission or triaging a class set - it mirrors the approach in our guide to detecting AI in GCSE Business Studies coursework.
Privacy protections depend on your account tier. On individual teacher accounts, which do not include detection, student work is redacted in the browser and never stored on any server. School accounts, where AI detection runs, operate under a signed Data Processing Agreement (DPA): pupils' work is held with their real names, uploaded scans are removed when a run archives (90 days by default, adjustable by the school from 7 to 365 days), and pupil records follow the school's retention window and are wiped at offboarding. Across all tiers, student submissions are never used to train AI models.
School accounts include unlimited AI detection checks alongside essay and coursework marking in every teacher seat, and a department can start with a free school trial. Details are on the GradeOrbit for Schools page.