Meningkatkan Integritas Akademik: Pelatihan AI Bagi Mahasiswa Baru dalam Mendeteksi dan Mengurangi Plagiasi Artikel Ilmiah

Abstract

Technological development Academic integrity is the main foundation in the development of science, but cases of plagiarism in scientific articles remain a serious challenge among students, especially during the transition period of higher education. This scientific article aims to develop and implement an artificial intelligence (AI)-based training program for new students to improve their understanding and skills in detecting and reducing plagiarism practices, and measure the effectiveness of the training in increasing students' awareness, knowledge, and technical abilities in producing original scientific works. This study uses a quasi-experimental method with a one-group pretest-posttest design approach. The results showed significant improvements in all measured aspects: conceptual knowledge of plagiarism increased by 45%, the ability to identify potential plagiarism in text increased by 62%, and the similarity index in student assignment articles decreased on average from 32% to 11%. The study's conclusions confirm that structured AI training effectively improves the academic integrity of freshmen by equipping them not merely as passive users of detection tools, but as intelligent and ethical writers capable of leveraging AI to strengthen the originality of their work.