The two-lane road to hell is paved with good intentions
Why an all-or-none approach to generative AI, integrity, and assessment is insupportable
Bibliographic Data
| ID | 21556263 |
|---|---|
| Authors | Guy J Curtis (0000-0002-4174-6955, The University of Western Australia, corresponding author) |
| Year | 2025 |
| Volume | 44 |
| Issue | 8 |
| Pages | 2151-2158 |
| Publication date | 2025-11-17 |
| Open Access | Yes |
| Type | ARTICLE |
| Venue | Higher Education Research & Development (JOURNAL) |
| Journal identifiers | ISSN: 0729-4360 • E-ISSN: 1469-8366 |
| Publisher | Informa UK Limited (PUBLISHER • GB) |
| DOI | 10.1080/07294360.2025.2476516 |
| OpenAlex | W4408644569 |
| Language | EN |
| Citations received | 7 |
| References cited | 18 |
A ‘two-lane’ (All-or-None) approach to the use of generative artificial intelligence (genAI) is the idea that there should be two categories of assessments in higher education: Lane 1/None: where the use of genAI is prohibited, and Lane 2/All: where any use of genAI is permitted. This idea has been thoughtfully detailed and continues to be debated. Although this idea is generally well-intentioned, in this comment piece I argue that, if implemented, it will promote an impoverished approach to education and educational assessment. One argument often invoked in favour of an All-or-None approach is that genAI use may sometimes be undetectable. Contract cheating (e.g., students outsourcing assessments to ghostwriters) is sometimes undetectable, yet an argument that there should be an All-or-None approach permitting contract cheating in some assessments is clearly absurd. An All-or-None approach to genAI and assessment is also absurd. A middle lane, where genAI use in assessments is allowed with some limitations, is essential
Engineering ethics · Epistemology · Generative grammar · Mathematics education · Academic integrity and plagiarism · Artificial Intelligence in Healthcare and Education · Computer Science · Engineering · Ethics and Social Impacts of AI · Philosophy · Psychology · Artificial Intelligence
Designing assessments in the generative AI era
Cognitive offloading, critical pedagogy, and connectivist learning shape academic responses to generative AI in higher education
ParadAIse L0st
AI literacy won’t save us
‘If GAI has finished her job, why would I need her?’
Dialogic Intelligence
AI and Assessment
Testing of detection tools for AI-generated text
‘Where’s the line? It’s an absurd line’
The educational integrity enforcement pyramid
Developing evaluative judgement for a time of generative artificial intelligence
Validity matters more than cheating
Using Digital Forensic Techniques to Identify Contract Cheating
Detection of GPT-4 Generated Text in Higher Education
Why is Cheating Wrong
| Citations per year | 7 |
|---|---|
| Citation span | 2025 - 2026 (2) |
| Highly cited | No |