News
Introducing Cadmus Oral Assessment: Scalable assurance for an AI‑enabled world

Share with colleagues
Download the full Case Study
Take an in-depth look at how educators have used Cadmus to better deliver assessments.
Cadmus is excited to announce the upcoming release of Cadmus Oral Assessment, a new solution designed to help universities deliver secure, defensible assessment in the age of generative AI.
Cadmus Oral Assessment introduces a new category of secure assessment: precision oral assurance — structured moments embedded directly into existing assessment journeys to strengthen evidence of learning. This is delivered through a phased approach: starting with a practical, low-friction way to introduce oral assessment today, and evolving into more targeted, contextualised assurance developed in partnership with universities to work in real institutional practice.
The first release of Cadmus Oral Assessment will be available from April/May 2026, following a co‑design program with leading universities in Australia, and the UK.
Why Cadmus Oral Assessments
Written assessment remains foundational to higher education. It develops critical thinking, research literacy, synthesis, and disciplinary argumentation.
But generative AI has changed how written work is produced.
Today, universities face a growing assurance gap: written submissions remain essential for learning, but may no longer provide sufficient evidence of understanding on their own.
Regulators are increasingly calling for assessment redesign — not simply more detection or surveillance — to protect validity, defensibility, and institutional trust.
Oral assessment remains one of the most trusted ways to validate reasoning and judgement — but traditionally they do not scale. They are time-intensive, difficult to schedule, inconsistent to moderate, and often impractical in large cohorts.
At the same time, many institutions have experimented with video-based responses and skills recording tools. While useful for reflection and employability, these approaches are not designed to deliver assessment assurance, nor to integrate into marking, moderation, and decision-making workflows.
Cadmus resolves this trade-off by delivering the assurance strength of oral assessment with the scalability, consistency, and workflow integration required for real university conditions.
Many other video-based tools have evolve from use cases such as:
- Skills demonstration and presentation practice
- Coaching and reflection
- Authenticity inferred from surveillance behavior and performance cues
Cadmus Oral Assessment was designed from day one for assessment assurance and learning validity:
- Embedded into marking and moderation workflows
- Aligned to rubrics and learning outcomes
- Designed to assess reasoning, disciplinary judgement and application of knowledge
“Cadmus Oral Assessment represents a major step forward moving secure assessment beyond surveillance and detection, and toward listening to students explain what they know. This is about protecting the integrity of qualifications in an AI-enabled world, without compromising pedagogy, trust, or student experience.”
What is Cadmus Oral Assessment
It is a learning‑centred, educator‑controlled oral assurance layer embedded directly into the Cadmus assessment platform.
Rather than introducing blanket oral exams or surveillance‑led controls, Cadmus enables:
- Targeted oral assessment moments only where explanation and judgement matter most enabling educators to make student thinking visible, not just measurable
- Oral explanation anchored directly to the student’s own written work and rubric
- Consistent, reviewable evidence embedded into marking and moderation workflows
The result is direct evidence of learning — without additional surveillance, unsustainable workload, or curriculum disruption.
Unlike tools designed primarily for skills demonstration, presentation practice, or reflective video responses, Cadmus Oral Assessments is purpose-built for academic assurance. It does not infer authenticity from behavioural signals — it captures reasoned explanation aligned to assessment criteria.
“We designed Cadmus Oral Assessment around one simple principle: credible assurance comes from listening to students explain their thinking, not from scoring signals or automating judgement. By grounding each oral moment in the student’s own work, capturing real-time reasoning, and keeping educators fully in control, this product delivers oral assessment that is authentic, scalable, and defensible in real university conditions.”
How Cadmus Oral Assessments works
Cadmus Oral Assessment integrates oral validation seamlessly into existing assessment workflows allowing institutions to scale.
Phase 1: Structured asynchronous oral assessment (Available April/May 2026)
The first release introduces a new assessment capability within Cadmus that enables scalable, low-friction oral assessment.
Educators can:
- Design structured Oral Assessment questions aligned to learning outcomes
- Deliver multiple video response questions within the same assessment
- Capture in-the-moment video responses under educator defined conditions
- Mark and apply rubric‑linked feedback using existing workflows and rubrics at scale and for audit and review
This enables universities to:
- Introduce oral assurance in a practical, proportionate way
- Preserve the spontaneity of a real-world Oral Assessment
- Validate student reasoning and understanding
- Pilot a scalable oral assessment solution without requiring full assessment redesign
Importantly, Phase 1 is designed to enable institutions to adopt a solution in a way that is scalable, explainable, and easy to implement.
What This Unlocks Next
As institutions begin to embed oral assessment into their workflows, Cadmus Oral Assessment is designed to evolve enabling more targeted, contextualised assurance across the assessment journey.
Phase 2: Contextualised Oral Assessment (Late 2026/2027)
Future phases will introduce optional AI-assisted capabilities to help educators scale and deepen oral assessment in a way that remains transparent and educator-led. Cadmus is exploring more personalised prompts based on a student's prior written work, rubrics, and educator prompts.
Phase 2 capabilities are currently being co-developed in partnership with universities, ensuring they reflect real academic practice and maintain transparency, control, and defensibility.
Cadmus AI will support educators in designing and scaling Oral Assessment experiences — while educators remain fully in control of prompts, marking, and outcomes. Cadmus AI is not intended to not mark, grade, or determine assessment outcomes.
Integrity by Design
Cadmus Oral Assessment strengthens academic integrity through assessment design.
- Oral Assessment moments preserve spontaneity and real-time reasoning
- Oral explanation is grounded in the student’s own work
- Students must justify and apply understanding in the moment
This approach reduces the effectiveness of external assistance and surfaces authentic understanding through explanation — not monitoring.
Built with universities, for real academic practice
Cadmus Oral Assessment is being developed through a phased, co-build approach with leading universities — ensuring it works in real academic conditions from day one, while continuing to evolve over time.
Phase 1 (April 2026) enables scalable learning validation through short, structured oral assessment moments.
The next phase of Cadmus Oral Assessment will expand to support more contextualised, personalised assurance — helping institutions strengthen confidence in assessment outcomes without increasing workload or complexity.
This approach ensures universities can start with a practical, proven model today, and scale into more advanced assurance over time.
Learn more
Cadmus Oral Assessments represents the next evolution of secure assessment — moving from watching students watching students perform to listening to them explain their thinking.
To learn more or participate in an early Pilot, contact the Cadmus team today.
Category
Product update
Company
More News
Load more
Academic Integrity
Leadership
Why universities agree with design-led integrity—but still struggle to move beyond detection
In this piece, Nick Bareham explores the gap between institutional belief and purchasing behaviour—and why procurement inertia, not lack of awareness, is slowing the shift to design-led integrity. From hidden reputational risks to the coming platform transition in assessment, this is a clear-eyed look at what’s really holding the sector back—and what it will take to move forward.
Nick Bareham, Chief Growth Officer, Cadmus
2026-04-27

Student Success
Using data to identify at-risk students early
A practical piece for educators on identifying at-risk students earlier by looking beyond final submissions. It shows how process-level signals, such as engagement patterns, drafting behaviour, and time-on-task, provide a clearer view of student progress, and how to act on those signals through early check-ins and more targeted support.
Cadmus
2026-04-20

Academic Integrity
Leadership
What a design-led model of academic integrity actually looks like in practice
In this piece, Cadmus Co-CEO Brigitte Elliott puts language to a growing shift in the sector: from treating academic integrity as a compliance challenge to understanding it as a function of assessment design. Drawing on decades of research, she outlines how institutions can move beyond detection-led responses and build systems that produce verifiable learning by design.
Brigitte Elliott, Co-CEO, Cadmus
2026-04-13