Sean Jernard Parker
Northcentral/National University, United States

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

Disclosure over Detection in AI-Assisted Academic Writing Sean Jernard Parker
International Journal of Educational Development and Innovation Vol 7, No 1 (2026): August
Publisher : Universitas Negeri Makassar

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26858/ijedi.v7i1.83716

Abstract

Generative artificial intelligence has changed academic writing by creating new uncertainty about authorship, assistance, assessment,and integrity. Many institutional responses rely on prohibition or detection, yet these approaches often occur after students have already acted without clear guidance. This conceptual paper proposes the Academic Integrity Disclosure Framework (AIDF), a values-based model for governing AI-assisted writing through transparency rather than surveillance. The framework was developed through a literature-informed synthesis of academic integrity scholarship, current research on generative AI in higher education, and the values of honesty, trust, fairness, respect, responsibility, and courage. The analysis shows that AI use exists on a continuum, that detection tools cannot adequately explain student intent or context, and that disclosure can better support ethical judgment before assessment occurs. AIDF organizes responsibility across expectation-setting, student disclosure, faculty interpretation, and institutional alignment. The paper concludes that disclosure-based integrity practices can help institutions protect learning outcomes while supporting fair, transparent, and educative responses to AI-assisted writing.