If you use ACJ to mark a task designed for a rubric, you aren't testing the technology, you are testing its ability to mimic an entirely different system. To move forward, we need a different starting point across six key areas.
Generative AI has broken one of higher education’s quiet assumptions: that a polished essay is a reliable proxy for student thinking. When tools can generate fluent academic prose on demand, we can no longer treat the final product as straightforward evidence of cognitive effort or authorship. The question for universities is no longer, “How can we prove this text wasn’t written by AI?” but “How can we design assessments where AI cannot replace the student’s contribution, only support it?”
At RM Compare, we believe that the true value of Comparative Judgement isn't just found in the final rank order (the product), but in the cognitive journey students take to get there (the process). Today, we are excited to share an experimental piece of work: the Learning Progress Dashboard.
The government’s post‑16 Level 3 and below reforms are no longer abstract policy; they are now a concrete redesign of the 16–19 landscape. A Levels, T Levels, new V Levels and two reformed Level 2 pathways will replace a crowded field of overlapping qualifications. In that world, assessment quality, standards and progression evidence stop being technical details and become existential questions for awarding organisations and providers.
For years, RM Compare has helped teachers, exam boards, universities and recruiters turn thousands of “which is better?” decisions into fair, reliable rankings of complex work. What began as a powerful way to compare pieces of work has now grown into something bigger: a complete ecosystem for creating, managing, and using gold‑standard judgements at scale. Today we’re introducing RM Compare | Live, Studio, Hub – three modules that work together as a continuous flywheel.
For this edition we have produced a couple of series focusing on two critical considerations in the world of assessment, and education more generally. The 'AI World' is moving fast and finding time for necessary thinking and consideration has never been more important.
Over this series we’ve built an argument step by step: Different curriculum models (content, product, process) need different assessment logics. Curriculum for Wales (CfW), as illuminated by Camau i’r Dyfodol, is best understood as a process curriculum. Adaptive Comparative Judgement (ACJ), via RM Compare, is an agnostic engine: it can serve any of these models depending on how it’s configured.
So far in this series we’ve: mapped three curriculum models (content, product, process) and their aligned assessments; seen how Camau i’r Dyfodol helped Wales understand Curriculum for Wales (CfW) as a process curriculum; and explored how Adaptive Comparative Judgement (ACJ), via RM Compare, is an agnostic engine that can serve any of these models depending on how it’s configured. In this post, we look at what happens when those pieces don’t line up
The core idea of this post is simple but easy to overlook: ACJ is an engine, not an ideology. It doesn’t come with “product” or “process” baked in. It will faithfully amplify whichever conception of quality and progression you plug into it. That makes it powerful – and dangerous – in systems where curriculum and assessment are not yet aligned.
In this post, we zoom in on Wales. The Camau i’r Dyfodol project offers a rare thing: a system‑wide, theoretically grounded attempt to help teachers understand a new national curriculum and to see what happens when they work explicitly with a process model. It also gives us a preview of what goes wrong when that model isn’t shared or isn’t matched by the wider assessment system.
When curriculum reform “fails”, we often blame teachers, resources, or “implementation”. But very often the real problem is simpler: the curriculum and the assessment system are speaking different languages. Before we can talk about Adaptive Comparative Judgement (ACJ) or RM Compare, we need a shared map of the curriculum landscape – and a clear view of the assessment that genuinely fits each terrain.
A City of London awarding organisation recently found itself in a strange position. Year after year, candidates were passing a respected Level 6/7 qualification. The statistics looked healthy, the quality assurance paperwork was in order and the exam board could point to detailed rubrics and grade descriptors. Yet employers were telling a different story.
What began as a powerful way to compare pieces of work is now a complete, production‑ready layer of judgement infrastructure that sits underneath your existing systems and makes subjective assessment fair, fast, and repeatable at any scale. We’ve reached the point Geoffrey Moore calls the “whole product”: not just the clever core technology, but everything wrapped around it that a mainstream organisation needs to trust it with real‑world stakes.