In 2025, Jisc launched the AI in Marking and Feedback Pilot, a year-long initiative bringing together colleges and universities to explore whether AI can meaningfully reduce marking and feedback workload in a way that is acceptable to key stakeholders.
The project spans two strands — tools designed specifically for educational purposes (these were Graide, Keath and TeacherMatic), and general-purpose AI tools (such as ChatGPT, Gemini, and Copilot), where the project focuses on custom assistants.
Across the initial months of the pilot, insights have been collected via regular community sessions, feedback forms and one-to-one interactions. As common themes appear, we’d like to share insights with the wider Jisc membership, allowing learnings to translate to direct value for the sector.
As such, we’re publishing this series of blogs, which we hope will give you a useful window into what the pilot has revealed about the role of AI within marking and feedback. You can read all the blogs in this series by following the links on this page.
Formative First
One of the strongest messages emerging from the AI in Marking and Feedback Pilot is that formative assessment is the best place to start. Regardless of the exact platforms being piloted, formative contexts consistently prove to be where AI delivers the most value with the least risk. This may seem like an obvious conclusion, but it is worth unpacking the reasons behind it.
With formative assessment, the emphasis is on delivering rich, actionable feedback to students in a timely manner — feedback that can be “fed forward” into their future work. Ensuring that this feedback is clear, detailed and consistent is often the central goal.
For summative assessment, by contrast, the integrity of the grade or mark is the primary concern, with feedback often serving the secondary role of explaining or justifying that mark.
These differences shape how appropriate AI is in each context.
In formative settings, the productivity gains offered by AI can translate directly into a better learning experience. Feedback can be generated and returned more quickly, meaning that students receive guidance while their learning is still in progress rather than weeks later when the opportunity to act on it has passed.
At the same time, AI systems can generate feedback that is rich in detail yet structured in a consistent way. When guided by a rubric, the AI can systematically analyse how a student’s work aligns with each criterion, helping ensure that key aspects of the task are addressed in the feedback.
Some participants in the pilot have raised questions about the accuracy and consistency of AI-generated marking. In particular, several institutions observed that running the same piece of work through an AI system multiple times can produce slightly different outputs. However, it is worth noting that variation is not unique to AI; different educators can also arrive at different judgements. In many institutions, existing moderation processes already exist to smooth out such differences.
From a student experience perspective, utilising AI within formative assessments also tends to cause less anxiety – particularly within the context of a pilot. Several institutions reported that students were more accepting of AI when it was framed as a learning companion rather than an automated assessor. When the purpose of the feedback is clearly developmental, students are often more open to experimenting with new approaches.
Some institutions have also used the efficiencies created by AI to expand the number of formative opportunities available to students. By reducing the time required to produce detailed feedback, educators have been able to introduce additional formative exercises, giving students more chances to practise, reflect and improve before summative assessments take place.
Despite being lower stakes, however, formative assessment is not a “no stakes” use case. Feedback still plays an important role in shaping how students develop their skills and understanding. Misleading or poorly calibrated feedback could therefore influence student learning in unhelpful ways. For this reason, moderation processes and human oversight remain important even in formative contexts.
When it comes to summative assessment, concerns around accuracy, transparency and fairness become more prominent, and many institutions remain cautious about allowing AI to play a significant role in determining final marks.
The experiences of the pilot therefore suggest a practical pathway for institutions exploring AI in assessment. Beginning with formative assessment allows educators to experiment with tools, refine workflows and build confidence among both staff and students. Over time, these experiences can inform more considered discussions about whether — and how — AI might play a role in higher-stakes contexts.
Find out more by visiting our Artificial Intelligence page to explore publications and resources, learn more about our communities and sign up for our AI Literacy training.
For regular updates from the team sign up to our mailing list.
Get in touch with the team directly at AI@jisc.ac.uk
