evidoria

← Back to browse

Good practice Imported

AI Assessment Scale (AIAS) — Reshaping GenAI-Era Exams at British University Vietnam, Hanoi

Vietnam · Hanoi · See the Vietnam profile · See the Hanoi profile

Evidence: Observational / pre–post Top 17% 67/100 · Ask Evidence Copilot about this practice

British University Vietnam piloted the AI Assessment Scale, a 5-level framework for permitted GenAI use in coursework; the peer-reviewed evaluation reported a 5.9% rise in attainment, a 33.3% rise in module pass rates, and fewer GenAI misconduct cases.

5.9 %
Increase in student attainment (2023-2024)
33.3 %
Increase in module pass rates (2023-2024)
AI Assessment Scale (AIAS) — Reshaping GenAI-Era Exams at British University Vietnam, Hanoi

Details

Maturity
Scaling
Promoter
British University Vietnam (BUV), with Deakin University and James Cook University Singapore
Period
2023–2024
Keywords
higher education, academic integrity, generative AI assessment policy

Context

Facing a wave of ChatGPT-related academic-misconduct cases after 2022, British University Vietnam (BUV), a roughly 2,500-student private university in Hanoi, worked with researchers from Deakin University and James Cook University Singapore to design the AI Assessment Scale (AIAS).

Objectives

Give instructors a five-level scale (Level 1 'No AI' to Level 5 'Full AI') for each assessment so permitted generative-AI use is explicitly scoped rather than banned outright or left ambiguous.

Activities

BUV piloted the AIAS across its modules and published the results in the peer-reviewed Australasian Journal of Educational Technology (AJET) in 2024.

Results

The pilot reported a 5.9% increase in student attainment across the university, a 33.3% increase in module pass rates, and a significant reduction in GenAI-related misconduct cases, alongside qualitatively reported gains in student engagement with transparently-used AI tools.

Conclusions

As a single-institution before/after pilot rather than a randomised trial, the reported gains have not been isolated from other concurrent changes at the university; the authors frame this as an initial implementation study rather than a definitive causal evaluation.

Implementation
Indicative cost
Low (< €50k)
Time to results
Short (< 1 year)
Staffing & skills
BUV instructors redesigning assessments, Researchers from Deakin University and James Cook University Singapore

Conditions for success

  • Instructor buy-in to redesign assessments around explicit AI-use levels
  • A simple, clearly communicable 5-level scale
  • Peer-reviewed publication enabling other institutions to adopt the framework

Common failure modes

  • No control group, so gains cannot be isolated from other concurrent institutional changes
  • The framework depends on per-assessment instructor judgement, which could be applied inconsistently across modules

Where it fits

Governance type
single private university
Scale
institution-wide pilot (~2,500 students)
Income level
lower-middle-income (Vietnam), private international institution

Replication kit

Reusable artefacts from this practice — as published by their sources.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Attachments

Similar practices you may find useful