Gray Areas and Alternative Resolutions on the Use of AI on Academic Assessments among ICT, TLE, and Language Educators

Abstract

The expansion of generative artificial intelligence (GenAI), including large language models that produce text through probabilistic prediction and pattern recognition, has complicated conventional assumptions about authorship, originality, and academic integrity in educational assessment. While institutional discussions often focus on detection technologies, limited scholarship has examined how educators interpret and resolve ambiguous student AI use in actual classroom contexts. Anchored in Weick’s Sensemaking Theory, this qualitative study examined how educators construct meaning around AI-related “gray areas” in assessment and how these interpretations shape instructional responses. Twenty (20) higher education educators from Information and Communications Technology (ICT), Technology and Livelihood Education (TLE), and Language disciplines in the Philippines participated in semi-structured interviews selected through purposive sampling. Ethical clearance was secured from the Institutional Review Board, and informed consent was obtained from all participants. Data were examined through Reflexive Thematic Analysis, involving iterative coding, reflexive journaling, peer debriefing, and thematic mapping to ensure analytic transparency and coherence. The findings demonstrated that educators’ interpretations were structured through three interconnected sensemaking processes: identity construction, enactment, and retrospection. Participants described tensions surrounding authorship verification, voice mismatch, process invisibility, and overreliance on AI detection tools. Rather than defaulting to surveillance-oriented enforcement, educators described adaptive strategies such as dialogic validation, staged drafting, oral defense, and contextual evaluation of learning evidence. This study advances a process-oriented explanation of how educators navigate AI-induced ambiguity in assessment settings. The integration of technological grounding with sensemaking constructs extends current discourse beyond descriptive concerns of misconduct and offers theoretically informed directions for institutional governance, professional development, and assessment redesign in AI-integrated academic environments.