vix.ing · top · new · best · stats · spec

Reasonable doubt about readiness: lessons from using MarkGPT (a custom GPT) to grade and provide feedback on law assessments

2026/07/29 by Simon Sneddon
Medicine · Social Sciences · #Artificial Intelligence in Healthcare and Education #Artificial Intelligence in Law #Ethics and Social Impacts of AI

paper · pdf · doi:10.1080/03069400.2026.2684407

openalex publication_date 2026/07/29 · openalex created_date 2026/07/30 · openalex updated_date 2026/07/31

Abstract

The emergence of Open AI’s ChatGPT in 2022 triggered considerable debate among academics both within law schools and widely across higher education. Initially, debate focused on students’ use of ChatGPT in creating their work. However, part of the discussion moved to academic use of Generative AI (GenAI), and one element of this was its use to grade student work and provide feedback. For some, GenAI marking has potential to be quicker and unbiased. For others, GenAI marking is part of the insidious creep of technology, risking academic redundancies. There is relatively little published material on GenAI and law essays generally, and nothing on using GenAI to mark and give feedback on law essays. This article emerges from the MarkGPT pilot project, which used a custom-built GPT to grade and give feedback on student work completed the previous year. The project compared the GenAI marking and feedback with the moderated grades from the module teams. The project concluded that, as of summer 2025, ChatGPT lacks the capacity for consistency (with itself, and with human markers), and accuracy in grading, but that a thematic analysis of the feedback showed significant similarities with the human markers. ChatGPT then, is not quite there. Yet.

Related