Artificial Intelligence and the Erosion of Academic Integrity

The proliferation of large language model (LLM) applications – most prominently ChatGPT – within Indian higher education institutions has precipitated a fundamental reconfiguration of academic work habits, assessment practices, and the cultivation of original scholarly thought. The present empirical investigation examines the relationship between AI tool adoption and academic integrity behaviour among students enrolled at Govt. L.B.S. P.G. College, Sironj, a semi-urban public institution in Vidisha district, Madhya Pradesh, with a total enrolment of 3,500. Employing purposive sampling, 50 student respondents (25 male, 25 female) were surveyed through a structured interview schedule, supplemented by 28 secondary sources. Data were subjected to frequency-percentage analysis across eight tabulated variables. Principal findings establish that: (i) 72% of respondents have submitted AI-generated content in academic assignments at least once; (ii) 68% acknowledge that they rarely or never verify AI-generated claims independently; (iii) 64% report declining confidence in constructing original arguments without AI assistance; (iv) 60% believe their institutions lack clear, enforceable AI usage policies; (v) 74% support formal digital ethics training as a mandatory curriculum component; and (vi) 66% recognise a measurable deterioration in their capacity to sustain extended analytical reasoning unaided. The study concludes that unregulated AI adoption in semi-urban colleges systematically undermines academic integrity norms, original scholarship, and critical cognitive development. Urgent recommendations include institutional AI governance frameworks, redesigned assessments, and mandatory digital ethics education.