(Image credit: Authors using CoPilot)
Introduction
Academics worldwide have been forced to take a hard look at their assessment practices the context of artificial intelligence (AI). For many, oral assessment has been considered as a relatively secure and robust option, although there are ever-present concerns about the security of even this form of assessment due to the emergence of AI wearables (Corbin, 2026). Oral examinations are not a new form of assessment. Their origins can be traced back to the fifth century, when Greek philosophers such as Socrates used oral questioning to challenge individuals, stimulate deeper thinking, and promote critical reflection. By the sixteenth century, examinations, viva voce, at Cambridge and Oxford were conducted orally, in Latin, and in public (Dobson, 2018).
In Ireland, the potential value of oral assessment has been further highlighted in the recently published Higher Education Authority Generative AI in Higher Education Teaching & Learning: Policy Framework (2025). The framework provides guidance to universities on the integration of AI while upholding academic integrity, transparency, and human oversight. One of the recommendations within the framework suggests that an “oral assessment safeguard” is a possible avenue for verifying authorship, when there are doubts as to the integrity of a student’s work. It proposed that educators should be able to invite students to explain or defend their work directly, with the outcome taking precedence over written submissions where needed (Higher Education Authority, 2025). This unambiguous directive is quite helpful for those in academic integrity units or similar functions in higher education organisations faced with offering institutional level academic integrity policies or guidance on assessment practices. Yet, when the time came for us to mould this recommendation into our own institutional academic integrity policy, it rapidly became apparent that the devil was in the detail. We hope that our reflections below will assist others facing a similar challenge.
Is the term ‘assessment’ appropriate?
We decided instead on the term live review.
This is because we are not suggesting that an assessment is conducted in the way a student would typically understand that term. Rather, the purpose of the process is to verify ownership of the artefact. The live review terminology better reflects the intended function of the activity: a live, interactive process through which the student discusses or demonstrates their work and demonstrates familiarity with its development and content. The emphasis is therefore not on reassessing the artefact, but on confirming authenticity, authorship, and engagement with the work.
The proposed live review differs from a traditional viva voce or oral assessment. It is not designed as a summative judgement of student performance, nor is it intended to function as a separate oral examination. It draws, however, on some of the strengths of oral interaction, such as immediacy, dialogue, and the ability to probe understanding, while using these features to support confidence in the integrity and ownership of submitted work.
How can we ensure a live review provides assurances relating to authorship and not inadvertently sanction a student who is simply less proficient?
The structure and focus of the interaction between the student and educator is key. The goal is not an additional assessment but rather an investigatory process. Thus, the focus shifts from the topic of the submitted piece to the process of its production, specifically the decisions the student took during the composition. Questions about the choice of methodology or the selection of specific argument structures allow the student to expand on their choices. Importantly, while a review prohibits novel supplementary questions about the topic, counterfactual questions related to the assignment itself can provide similarly comprehensive safeguards. Questions such "what would the consequences have been if your idea had been X rather Y? How would your conclusion change if this assumption were removed?” The purpose is not to expect identical performance across assessment modalities. Many students may write better than they speak, they may suffer from anxiety related to discursive tasks or may struggle if not speaking their native language. However, assessors can look for consistency in conceptual understanding, familiarity with sources awareness of limitations, knowledge of methods used. In this way a discrepancy becomes meaningful only when it is substantial and systematic, not merely a difference in communication skill.
Is the introduction of an institution wide policy statement encouraging an oral assessment safeguard sending out the wrong message? Should priority be given to assessment redesign in order to uphold academic integrity?
There is a need for both.
Production of a written artefact can be an invaluable for the learning process. This has been well described by Corbin (2026) as a ‘process of becoming’, promotes self-regulatory learning and critical thinking gained through the iterative approach to writing. Many academics may therefore wish to preserve this aspect within the modules they oversee and the prospect of a live review may help to provide assurances of veracity. However, realistically, this approach is not an entirely certain method of detecting academic misconduct and indeed the workload associated (marking of the submission and conducting a live review) may be quite laborious. Therefore, if the key concern is ensuring the integrity of the assessment, it may be necessary to consider assessment redesign as a more robust approach. As we view it, the best course of action is dependent on the purpose of the assessment, which is directly linked to the knowledge, skills or competencies under examination.
References
Corbin, T., Sharpe, S., & Dawson, P. (2026). On AI glasses and wearable AI in assessment. Assessment & Evaluation in Higher Education, 1–17. https://doi.org/10.1080/02602938.2026.2661367
Corbin, T., Walton, J., Bannister, P., & Deranty, J.-P. (2026). On the essay in a time of GenAI. Educational Philosophy and Theory, 58(3), 198-210. https://doi.org/10.1080/00131857.2025.2572802
De Vita G., Case P. (2003). Rethinking the internationalisation agenda in UK higher education. Journal of Further and Higher Education, 27(4), 383–398. https://doi.org/10.1080/0309877032000128082
Dobson S. (2018). Assessing the viva in higher education: Chasing moments of truth. Springer International Publishing. https://doi.org/10.1007/978-3-319-64016-7
Higher Education Authority (2025) 'Generative AI in Higher Education Teaching & Learning Principles for Ethical AI Adoption'. Generative AI in Higher Education Teaching& Learning Principles for Ethical AI Adoption | Publications | Higher Education Authority
Silvia Benini is an Educational Developer in the Academic Integrity Unit at the University of Limerick.
Mary-Claire Kennedy is an Associate Professor and Academic Integrity Lead at the University of Limerick. She is Chair of the Irish National Academic Integrity Network (NAIN).
Fionn McGrath is and Educational Developer in the Academic Integrity Unit at the University of Limerick.
Reena Cole is Associate Vice President Academic Affairs at the University of Limerick
The authors' views are their own.
Thank you for being a member of ICAI. Not a member of ICAI yet? Check out the benefits of membership and consider joining us by visiting our membership page. Be part of something great!