Marking.ai and a general AI chatbot
Both will give you a mark. The difference is what the mark is measured against, and whether you can see why.
Can I just use ChatGPT to mark my class set?
You can get a mark from a general-purpose chatbot, but it marks against its own idea of a good answer rather than your mark scheme, it does not reliably show which words earned which mark, and it starts fresh on every script - so the standard drifts across a class set. Marking.ai extracts criteria from the guidelines you upload, cites the student's own words for every mark, and applies one configuration to the whole set.
Side by side
Capabilities that follow from how each tool works, rather than opinions about either.
| Feature | Marking.ai | General AI chatbot |
|---|---|---|
| Marks against your uploaded mark scheme | Yes | No |
| Criteria extracted per question before marking | Yes | No |
| Quotes the student's own words as evidence | Yes | No |
| Reads handwritten scripts | Yes | No |
| Same configuration across a whole class set | Yes | No |
| Teacher approval before a student sees anything | Yes | No |
| Question-level reporting after marking | Yes | No |
| Built for school procurement | Yes | No |
It reads your marking guide first
Marking.ai takes the marking guidelines or mark scheme you already use, extracts the questions and the criteria out of your own document, and marks each answer against them one criterion at a time. The standard being applied is the one you wrote down, and you can see it beside every mark.
- Your document, structured into questions and criteria
- Each answer marked criterion by criterion, not by overall impression
- The same standard applied to every script in the class
The difference is the mark scheme, not the model
This is not a claim that one model is cleverer than another. Both are language models and both can read an essay. What differs is what the marking is anchored to.
A chatbot is given an answer and asked for a mark, so it marks against whatever it understands good work to look like. Marking.ai is given your marking guidelines first, extracts the criteria for each question, and marks each answer against those criteria - returning the student's own words as the evidence for every mark it awards.
That is also why the second script gets the same treatment as the first. The configuration is attached to the assessment, not re-established in each conversation.
Every mark comes back with the words it rests on
A mark on its own is a verdict. Marking.ai returns the exact words in the student's answer that earned each mark and the page they were found on, so disagreeing takes seconds and changing a mark takes one edit. Nothing reaches a student until a teacher has approved it.
- The exact student text behind each mark, with its page number
- Change any mark, and the change is what stands
- A teacher approves before a student sees anything
Judge it against your own marking
Mark one set you have already marked by hand, and compare. No card required to start.
In short
- A chatbot marks against its own idea of a good answer; Marking.ai marks against the criteria in the guidelines you upload.
- Marking.ai quotes the student's own words as the evidence for each mark.
- One configuration applies to the whole class set, so the standard does not drift between scripts.
- A teacher approves every mark before a student sees it, in both cases - but only one of them makes that a step in the workflow.
Common questions
Isn't AI marking inaccurate?
Marking.ai marks to your marking guidelines and mark scheme criterion by criterion, and shows the evidence behind every mark - so it's accurate and auditable, not a black box. It stays consistent across a whole class, and a teacher reviews and approves every mark before it reaches a student: a faster first pass, not an unchecked final mark.
Will it replace or deskill teachers?
No. Marking.ai drafts; teachers decide. Nothing is released until a teacher approves it. It removes the repetitive first pass so teachers spend more time teaching and on the judgement only they can make.
Could I not just paste my mark scheme into a chatbot?
You can, and for one script it works reasonably well. It stops working across a set: the criteria have to be re-established every time, nothing holds you to the same interpretation on script thirty as on script one, and there is no record of which words earned which mark when a student or a parent asks.
Assessment terms explained
Plain-English definitions of the vocabulary used on this page.
- AI markingAI marking is the use of artificial intelligence to assess student work against defined criteria - a mark scheme or a teacher's own marking guidelines - and to draft marks and feedback for a teacher to review.
- Assessment for learningAssessment for learning is the practice of using evidence of what students currently understand to decide what to teach next, and to help students see what they need to do to improve.
- Assessment objectivesAssessment objectives are the categories of skill a qualification tests - commonly labelled AO1, AO2 and AO3 - each carrying a defined share of the available marks.
- Comparative judgementComparative judgement is an assessment method in which markers repeatedly choose the better of two pieces of work, and those paired decisions are combined statistically into a rank order.
- Criterion-referenced assessmentCriterion-referenced assessment judges work against defined standards of what a student should know or be able to do, rather than against how other students performed.
- ExemplarAn exemplar is a piece of work - often a real student answer - presented with its marks and commentary to show what a particular standard looks like in practice.
Judge it against your own marking
Mark one set you have already marked by hand, and compare. No card required to start.
