Investigating Lecturers’ Perceptions and Attitudes towards the Use of Artificial Intelligence in Teaching and Learning: Critical Empirical Evidence and a Case-Study Framework for the Higher Institute of Water Affairs, Ajilat, Libya

Abstract

Generative artificial intelligence (AI) is gaining acceptance in universities, but positive attitudes may coexist with uncertainty about accuracy, responsibility and technical safety. This article critically re-analyses verifiable empirical findings from Libyan higher education and recent UK surveys to develop a theoretically specified case-study framework for the Higher Institute of Water Affairs, Ajilat. It is a secondary-empirical synthesis, not a report of new lecturer interviews at the institute. A Libyan study of 60 EFL instructors reported behavioural intention towards AI-assisted lesson planning at M = 4.06/5 and verification behaviour at M = 4.40/5, with concern and intention positively associated (r = .420, p < .001); 68.3% of open-ended responses identified time saving and 48.3% over-reliance as a concern. A separate Benghazi study (n = 20) found broadly favourable AI perceptions coupled with self-taught rather than formal adoption. UK Jisc teaching-staff data (n = 1,765) reported AI use in teaching at 37% and availability of AI training at 23%. A 2026 UK undergraduate survey (n = 1,054) found 95% had used AI and 38% reported institutional provision of AI tools. The synthesis identifies “critical ambivalence”—willingness to use AI alongside scrutiny—as a plausible feature of responsible practice, not a deficiency. However, self-reported verification is not evidence of output correctness, and Libyan EFL findings cannot be generalised to high-stakes water-sector instruction. The paper presents an original evidence matrix, quantitative source diagrams, a task-risk taxonomy and a feasible Ajilat empirical protocol.