Generative artificial intelligence (GenAI) has emerged as an influential development in higher education, prompting renewed attention to assessment, academic integrity, and authorship. Although research on GenAI has expanded rapidly following the emergence of large language models, existing studies have often examined technological capabilities, pedagogical applications, and integrity concerns separately. Consequently, limited attention has been given to how patterns in the scholarly literature correspond with the experiences of students and instructors navigating GenAI in assessment contexts. This study addresses that gap through a complementary mixed-method design that combines a bibliometric analysis of 1,685 Scopus-indexed publications focused on academic integrity and assessment with qualitative interviews involving 48 participants (27 students and 21 instructors) from higher education institutions in the Cebu region of the Philippines. The bibliometric analysis identified a marked increase in scholarly output after 2022 and revealed three dominant areas of inquiry: detection and prevention strategies, pedagogical adaptation, and ethical governance. The qualitative findings showed that students commonly use GenAI to support learning, productivity, and academic tasks while negotiating unclear expectations regarding acceptable use. Instructors, meanwhile, increasingly respond by redesigning assessments, emphasizing process-oriented evaluation, authentic learning tasks, and performance-based approaches rather than relying exclusively on AI-detection technologies. Taken together, findings from the integrity- and assessment-focused Scopus corpus and the Cebu-based qualitative inquiry suggest that, within these contexts, GenAI is associated with evolving approaches to academic integrity and assessment that extend beyond concerns about academic misconduct alone. The findings underscore the importance of context-sensitive institutional policies, AI literacy initiatives, and assessment practices that promote transparency, accountability, and meaningful demonstrations of student learning in AI-enabled higher education. Given the study’s methodological scope, these findings should be interpreted within the context of the bibliometric corpus analyzed and the qualitative setting in the Philippines, while providing insights that may inform future research and practice in comparable higher education environments.