Skip to main content
AiGORA
Back to Q&A
TextQ&A
17

Can I use AI to help me with grading exams?

I guess this is too general as a question, a bit more nuanced here:

  1. Can I use AI to transcribe students' handwriting from their exams? Students from some countries write in a cursive style that I find very difficult to read, and generally most students' handwriting is terrible. Is it ok to use AI to help me decipher?
  2. Assuming that the answer to 1 is yes, now that I have all student answers digitally, is it ok to let AI group it in ways that make it easier for me to mark?
  3. Assuming the answer is still yes, can I ask AI to write feedback to students? That is so much faster than me doing it, and also it is more friendly than my tired self at 23:47 at night when they all get the same thing wrong...
  4. Should I tell students that I use AI on their exams and how?

Anon · 13 Nov 2026

Responses from the team

3 perspectives from the community

  • Kirsty Dunnett

    Although you lay out the aspects of your question as sequential, I think they're not quite so linear. I will, however, address the different aspects in order. 1. Unless you can be absolutely certain of data protection and that use for transcription is explicitly permitted, I would use for no more than individual words or at most clauses. Optical character recognition is done by pixel, and handwriting is a big challenge; multi-modal LLMs that are trained to handle images struggle, so the chances that a lot of errors will occur is high. A strong language dependence may also be expected, and the 'difficult' cursive handwriting will be particularly liable to LLM errors, so your biases are not overcome. See, e.g.,  https://arxiv.org/abs/2502.20295 and https://arxiv.org/abs/2503.15195 2: I have to ask: But why do you need the LLM to group responses? My undergraduate lecturers once told me that the way to mark scripts, whether homeworks or exams, quickly is to mark by question. One goes through the stack several times, but one marks much more quickly because one only has to remember the answer to the current question (or even question part). Let me also add that, in spring 2026, a colleague at an institution that is generally pretty 'pro-AI' experimented with using an LLM ('sandboxed') to mark student reports and abandoned the effort after a few hours: it was more frustration and work to get the LLM to provide the feedback he wanted than it was to just sit down and work through the reports as he had for the many years before 'generative AI' products could be claimed to be sufficient for such tasks. 3: Here is where I think LLMs can be helpful, though I am going to start with a warning. LLMs work on the 'norm'; the average, and is what automatically generated feedback will do (or, if marking against a model answer, not recognise equivalent approaches or different solutions). Do you really want your students to be dragged to some mediocre average? Or punished for taking a different approach and showing something equivalent to the required work? However, I can see that LLMs may help with one thing: creating feedback comments that you only need to edit a little. So you could start your marking knowing (from experience) the things that you tend to want to write something quite 'unfriendly' in response to (i.e., the things that frustrate you), and spend a few minutes developing some templates for you to adjust as necessary. Another strategy I would suggest here, and irrespective of whether you use LLMs at all, is to flick through the scripts before marking: if they are all getting the same thing wrong (and especially if they do it in the same way), you can develop a common, specific feedback. Depending how long and detailed that feedback is, you might also choose to not provide feedback on that particular question, instead marking it as right or wrong (if wrong, point to the common feedback) and instead provide a single feedback for everyone, pointing out that many students got it wrong in the same way. You will need to be very clear about redirecting students to the 'correct' solution. (If you use the same questions for several years, you could make a copy of a characteristically wrong response and annotate it, saving yourself the work in future years.) 4: You should definitely tell your students if you use LLMs for any part of the marking (or plan to). If you don't you will almost certainly destroy their trust in you (though it's no guarantee you won't destroy their trust in you if you tell them). If you require them to declare LLM usage or advise against it, and then use it yourself, especially without admitting it, you are also being dishonest and hypocritical. Would you have liked your lecturers to have treated your own work with such disdain?

    Read all posts that Kirsty has responded to →

  • Rachel Forsyth's profile photo

    Rachel Forsyth

    Here are a few suggestions to add to Kirsty's answer:

    If you use a university-purchased system (in EU/UK at least), and check in the settings that nothing is saved when you upload to the GenAI product, the data should not be used for training or saved in the manufacturer's system. Maybe you will get a better output from the system than your tired eyes, but it isn't guaranteed, so you'll still have to cross-check - it could end up taking more time. That doesn't mean I endorse it, you should definitely discuss with students what they think about it - they might be sympathetic in this situation, actually, but don't assume it without checking with them.

    Your tired eyes are the underlying problem here - so whatever you decide to do this year, put aside some time to change this assessment for next year:

    a) Change the format of the exam away from handwriting. Most of us don't practise sustained handwriting any more. Use MathLab if this is for problems. At Lund University, you can also use the Inspera software to have supplementary sheets scanned in with elements which need to be hand-drawn or -written, this might be easier to manage alongside most of the material being typed. b) Use GenAI to produce a good quality feedback bank before you start grading. Use your grading rubric and the assessment information to generate constructive, actionable comments. I explain how to do this in this open access chapter. Obviously you need to check them and pick the best ones, and adapt them to individual assignments as you grade, but it can save you time thinking of helpful comments while you are grading.

    PS if you don't have a good grading rubric, you are also making work for yourself while you grade - the same chapter suggests how you can use a GenAI product to help you make one, but you WILL need to edit it.

    Read all posts that Rachel has responded to →

  • Mirjam Glessmer's profile photo

    Mirjam Glessmer

    I would not advise to upload student work to AI without their knowledge and permission, and from talking with students, it typically matters to them a lot that their work is being assessed by a human, not a robot

    Read all posts that Mirjam has responded to →

Comments

Share your thoughts — comments are reviewed before they appear

No comments yet. Be the first!