While previous studies have explored students' use of different AI tools for academic purposes, studies that have specifically investigated students' use of ChatGPT for dishonest academic purposes in Nigeria are lacking. The consequence of this contextual and knowledge gap is a lack of specific understanding regarding students' engagement with ChatGPT for academic dishonesty in Nigerian tertiary institutions. This study addressed these gaps by examining students' perceptions, attitudes, and utilisation of the ChatGPT and determining the role of sex and age in these linkages. A sample of 4679 public university students participated in the study. Structural equation modelling and multigroup analysis were performed to test the conceptual model with the aid of SmartPLS 3. The results indicated that, regardless of sex or age, students with positive perceptions of ChatGPT were more prone to use it for dishonest academic purposes. The study noted a sex disparity in the direct impact of perception on ChatGPT use, which was particularly pronounced for female students. Significant age-related differences were observed, with a stronger effect observed for younger students. A negative direct effect of attitude on ChatGPT use for academic dishonesty was recorded, with attitude further serving as a significant negative mediator of the relationship between perception and ChatGPT use. This mediating effect was consistent across sexes but varied with age, being stronger among younger students than among their older counterparts. This study underscores the need to foster positive attitudes among younger students to counteract the appeal of using the ChatGPT for academic dishonesty.