QAA's AI assessment report says inconsistent practice is now a student experience risk
Updated Jul 10, 2026
QAA's new AI assessment report deserves immediate attention because it reframes AI as a student experience and quality problem, not only an integrity problem. On 9 July 2026, QAA published New research reveals the variability of policies, practices and student experience in the age of AI, summarising its latest report on AI and assessment. For teams responsible for student voice, the practical message is clear: if students meet different AI rules, different levels of staff confidence, and different assessment expectations inside the same programme, that is now a quality risk as well as a communications problem.
What has changed in QAA's AI assessment report
QAA's announcement sits alongside its new State of the Nation report, The perfect storm: AI, assessment and a sector under pressure, published on 7 July 2026. QAA says the report draws on staff roundtables, student focus groups, thematic analysis of QAA review reports from 2023 to 2026, and recent sector evidence on generative AI use. This is a sector-wide diagnostic, not a new regulatory rule or a survey methodology change. But it still matters because it spells out what QAA now sees as the live risk.
The central finding is not simply that AI use is growing. It is that practice is uneven inside institutions as well as between them. QAA says policies are being applied inconsistently across departments, programmes, modules, and individual tutors, creating confusion for staff and students and producing material differences in the learner experience. In the report summary, QAA highlights five areas of risk: assessment validity, parity of student experience, trust between staff and students, the development of foundational skills, and the pressure created by tight budgets and fast-changing tools.
"The sector has acted, but practice is uneven."
— QAA, The perfect storm: AI, assessment and a sector under pressure
QAA also sets out what it wants the sector to do next. The news announcement says institutions should invest in staff and student training, build student voice into AI policy and guidance from the outset, and make consistency of student experience a priority. On the State of the Nation page, QAA says it will respond through an AI in Assessment Community of Practice, a 2026-27 membership offer that treats GenAI as a cross-cutting theme, and a refresh of the Academic Integrity Charter during the forthcoming academic year. The signal for institutions is practical: QAA is not asking for more rhetoric about AI readiness. It is asking for clearer policy, clearer implementation, and clearer evidence that students understand the rules they are being asked to work within.
What this means for institutions
The first implication is that universities should audit variation at programme level, not just publish another institution-wide AI statement. If students encounter one set of expectations in a seminar, another in a module brief, and a third in marker behaviour, the issue is no longer only policy wording. It becomes a parity problem that can affect trust in assessment itself. Student Experience teams and quality professionals should therefore look for inconsistency across modules, schools, and delivery teams, especially where assessment redesign has moved quickly.
The second implication is that student feedback collection now needs to ask more precise questions. Generic prompts about whether students feel positive about AI will not tell teams enough. Institutions need to know whether students understood what AI use was permitted, whether guidance matched practice, whether staff responses were consistent, and whether students felt assessment still tested their own capability fairly. A short student comment analysis governance checklist is useful here because it helps teams document which evidence routes are in scope, who reviews them, and how conflicting signals are escalated.
The third implication is about evidence for quality assurance. QAA's emphasis on training, student voice, and consistency means universities will need a clearer line from student comment to institutional response. That includes showing where AI guidance was unclear, what was changed, and whether students experienced the revised approach more consistently afterwards. If that follow-through is weak, institutions may find that AI appears in feedback as a fairness and trust problem before it appears as an innovation success.
How student feedback analysis connects
This is exactly the kind of issue that open-text feedback exposes faster than headline scores do. Students will describe conflicting instructions across modules, vague wording about permitted AI use, different marker expectations, or the sense that one tutor encourages tools another treats with suspicion. A reproducible method such as our NSS open-text analysis methodology helps teams compare those themes across module evaluations, local pulse surveys, and representative channels without flattening them into one broad AI category.
It also sharpens the case against ad hoc generic LLM workflows when the output needs to support committee decisions or policy revision. If the institutional problem is inconsistency, the analysis method should not introduce more of it. Where universities need to compare large comment sets with a clearer audit trail, Student Voice Analytics is one practical option. The more important point is methodological: AI-related student voice evidence needs to be reviewable enough to show where variation sits, who is affected, and what changed in response.
FAQ
Q: What should institutions do now?
A: Start with a targeted audit of AI guidance across modules and schools. Check whether students are being told the same thing in assessment briefs, handbook wording, staff explanations, and academic integrity processes. Then use module evaluations, rep forums, and open-text survey routes to test whether students experienced the rules consistently in practice.
Q: What is the timeline and scope of the QAA change?
A: QAA published the report The perfect storm: AI, assessment and a sector under pressure on 7 July 2026 and the news announcement on 9 July 2026. This is a sector-wide QAA analysis rather than a new statutory rule, but QAA says it will respond through an AI in Assessment Community of Practice, a 2026-27 membership offer, and a refresh of the Academic Integrity Charter in the coming academic year.
Q: What is the broader implication for student voice?
A: Student voice is becoming one of the clearest ways to see where AI policy looks coherent on paper but inconsistent in day-to-day assessment. Universities that can compare comments on clarity, fairness, and trust across modules will be better placed to intervene before inconsistency turns into complaints, weak evidence, or external scrutiny.
References
[Quality Assurance Agency for Higher Education]: "New research reveals the variability of policies, practices and student experience in the age of AI" Published: 2026-07-09
[Quality Assurance Agency for Higher Education]: "The perfect storm: AI, assessment and a sector under pressure" Published: 2026-07-07
[Quality Assurance Agency for Higher Education]: "Academic Integrity Charter" Published: not stated
Request a walkthrough
Book a free Student Voice Analytics demo
See all-comment coverage, sector benchmarks, and reporting designed for OfS quality and NSS requirements.
- All-comment coverage with HE-tuned taxonomy and sentiment.
- Versioned outputs with TEF-ready reporting.
- Benchmarks and BI-ready exports for boards and Senate.
UK-hosted · No public LLM APIs · Same-day turnaround