Performance without capability: generative AI and the validity of assessment in higher education

Generative artificial intelligence (GenAI) has altered how assessment supports learning and evidences achievement in higher education. Early institutional responses concentrated on academic integrity, detection, and task redesign, while a rapidly expanding empirical literature documents the contingent benefits and trade-offs of AI-mediated learning. This conceptual paper develops an assessment-specific framework organised around two dimensions: the distribution of learning-relevant cognitive responsibility and the visibility of student process. GenAI can operate as an engagement amplifier when it supports cognitive and metacognitive activity while preserving student responsibility for the intellectual work that matters, or as a cognitive substitute when it displaces that work while enabling successful task completion. The framework distinguishes immediate AI-supported performance from durable capability, whether demonstrated independently or with GenAI, and connects this distinction to the validity of inferences drawn from assessment evidence. Rather than proposing a universal solution to what has been called the wicked problem of AI and assessment, the framework offers an evaluative lens for judging how particular assessment configurations distribute cognitive responsibility and make student thinking visible, and whether the resulting evidence can support defensible claims about achievement. This reframing informs assessment judgements without prescribing a single model for all higher education contexts.

Authors

Institutions

Publication Details

Journal
Assessment & Evaluation in Higher Education
Published
2026-11-18
DOI
https://doi.org/10.1080/02602938.2026.2764204
Primary Topic
Artificial Intelligence in Education
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Performance without capability: generative AI and the validity of assessment in higher education

Sean Kearney
4 links
Assessment & Evaluation in Higher Education
Artificial Intelligence in Education
article

Performance without capability: generative AI and the validity of assessment in higher education

Sean Kearney
article en

Abstract

Generative artificial intelligence (GenAI) has altered how assessment supports learning and evidences achievement in higher education. Early institutional responses concentrated on academic integrity, detection, and task redesign, while a rapidly expanding empirical literature documents the contingent benefits and trade-offs of AI-mediated learning. This conceptual paper develops an assessment-specific framework organised around two dimensions: the distribution of learning-relevant cognitive responsibility and the visibility of student process. GenAI can operate as an engagement amplifier when it supports cognitive and metacognitive activity while preserving student responsibility for the intellectual work that matters, or as a cognitive substitute when it displaces that work while enabling successful task completion. The framework distinguishes immediate AI-supported performance from durable capability, whether demonstrated independently or with GenAI, and connects this distinction to the validity of inferences drawn from assessment evidence. Rather than proposing a universal solution to what has been called the wicked problem of AI and assessment, the framework offers an evaluative lens for judging how particular assessment configurations distribute cognitive responsibility and make student thinking visible, and whether the resulting evidence can support defensible claims about achievement. This reframing informs assessment judgements without prescribing a single model for all higher education contexts.

Assessment & Evaluation in Higher Education
The University of Notre Dame Australia (AU)
Quality education
Artificial Intelligence in Education
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon