Introduction
We met on the 15th July 2025 for our final higher education community session of the academic year. The conversation centred on practical AI in teaching, assessment integrity, and what feedback students are telling us around employability. Below, we’ve pulled together the key highlights for anyone who couldn’t join us.
Teams Premium and Co-pilot
Several members noted that Microsoft Teams Premium now provides high‑quality meeting transcription (and recordings) at around £20 per user per year, making it a cheaper option than Co-pilot licences when you primarily need transcripts and are happy to view notes after the meeting. This could be a pragmatic route for institutions wanting reliable accessibility support without full generative AI features.
Agents
Some members reported early exploration of building Microsoft 365 agents. Licences allow internal deployment, but sharing beyond licensed users incurs additional credit costs, potentially unpredictable at scale. Worth prototyping, but these cost models will need scrutiny.
Employability
Student feedback gathered across institutions, and reinforced in Jisc’s Student Perceptions of AI survey, keeps placing employability at the top of the AI conversation.
- Students want to know how AI is being used in their courses so they can mirror realistic workplace practice.
- Guidance on ethical, effective use (CVs, applications, interview prep, workplace tools) is in demand.
- Where staff flag AI use transparently, students report greater trust and lower anxiety.
Academic Integrity and Detection
An item in the meeting Padlet on spotting AI use in student work started a wider discussion. Key takeaways:
What’s actually showing up in cases?
- colleagues have handled recent misconduct cases in which fabricated or unverifiable references were copied directly from AI outputs.
- Arguments that “this doesn’t sound like the student” are weak where institutions formally permit AI writing support; style shift alone doesn’t evidence misconduct.
Better conversations
- Focus on evidence you can trace: cited sources that exist; drafts, prompts, and iteration history; opportunities for students to explain their workflow.
- Document your process thoroughly. Recent sector guidance and appeals data suggest weak documentation undermines institutional decisions.
Assessment design still matters
- Give clear, practical guidance to students about acceptable AI use.
- Use a mix of assessment methods (in‑class, oral, practical, iterative) to reduce over‑reliance on text submissions.
- Ensure tasks assess the knowledge and skills learners should actually develop – the “can you still do the job if the power goes out?” test.
AI in marking and feedback
Survey findings shared in the session indicate many students remain averse to AI awarding marks. Several said they assume AI isn’t used unless institutions explicitly state otherwise.
At the same time, a pattern is emerging:
- Openness to AI‑assisted feedback on spelling, grammar, structure, and academic writing style – especially if it means quicker turnaround.
- Caution for content judgement: students want human academics involved in evaluating subject understanding.
What students say they value:
- Consistency across markers
- Faster, clearer formative feedback
- Transparency about when/where AI assists staff – provided a human signs off.
Next HE meet‑up
September 2025 (date TBC), Where we’ll workshop community priority areas for 2025/26.
Links shared during the call
Find out more by visiting our Artificial Intelligence page to view publications and resources, join us for events and discover what AI has to offer through our range of interactive online demos.
For regular updates from the team sign up to our mailing list.
Get in touch with the team directly at AI@jisc.ac.uk