Generative AI-Assisted Academic Writing in Higher Education EFL Contexts: A Critical Review of Learning Outcomes, Self-Regulated Learning, and Ethical Governance
Aug 2026· Journal of English language teaching and applied linguistics· Vol 8, pp. 01-11· 0 citations
TL;DR
It is argued that GenAI’s value cannot be judged by writing products alone but requires understanding the interplay among cognition, autonomy, and regulation.
Abstract
GenAI’s rapid proliferation has reshaped EFL academic writing, yet research remains fragmented. This review synthesizes 24 publications (2022–2026) on learning outcomes, self-regulated learning (SRL), and ethical governance. Three findings emerge: (1) positive effects on surface-level writing are confirmed but moderated by intervention duration, task complexity, learner proficiency, and evaluation methods; (2) GenAI’s impact on SRL depends on its role as a scaffold versus a substitute for learner agency, as illuminated by TAM, SCT, and SRL; (3) ethical governance is shifting from plagiarism policing to proactive AI literacy. Gaps persist in longitudinal and process-oriented designs, representation of underserved groups, and theory-driven pedagogy. The review argues that GenAI’s value cannot be judged by writing products alone but requires understanding the interplay among cognition, autonomy, and regulation. An ecologically-informed research agenda is proposed.
opportunities for self-regulation and risks of academic dependency. This narrative review aimed to synthesize how GenAI may support self-regulated learning (SRL), how intensive or uncritical use may promote cognitive offloading, and which contextual conditions appear to distinguish these outcomes. Ten peer-reviewed articles published between 2022 and 2026 were identified through Publish or Perish using Google Scholar and synthesized thematically. The evidence indicates that GenAI can support SRL by functioning as a learning tutor, assisting goal setting and planning, facilitating monitoring and self-evaluation, and improving learning efficiency. At the same time, overreliance on GenAI is associated in the reviewed literature with cognitive offloading, academic dependency, reduced cognitive effort, weaker critical evaluation, and diminished learner autonomy. The synthesis proposes four contextual boundary conditions - purpose of AI use, SRL capability, AI literacy, and intensity of use - that may shape whether GenAI acts primarily as a learning scaffold or as a substitute for cognitive engagement. The framework is conceptual and has not been empirically validated. Higher-education institutions should therefore combine GenAI access with pedagogical practices that strengthen planning, verification, reflection, critical thinking, and responsible AI literacy.
Generative Artificial Intelligence (GenAI) is rapidly reshaping higher education, especially in technology-oriented programs where critical thinking and complex problem solving are core outcomes. This study triangulates global and local evidence on performance/efficiency, usage, autonomy, critical-thinking engagement, and ethics by combining a systematized review informed by Kitchenham and structured using selected PRISMA 2020 elements (2020–2025; last search: May 2025; 49 studies; Scopus, ACM Digital Library, IEEE Xplore, and SpringerLink; not prospectively registered) with an anonymous survey of 302 computing and engineering students from a single university in Ecuador. The expert-reviewed instrument showed acceptable internal consistency for most scale-based dimensions (McDonald’s ω), whereas institutional and ethics-related items were analyzed primarily at the item level. Results showed near-universal academic GenAI use (96%), with 47% of students reporting weekly use and 26% daily use. Research-related work was the most frequent application (81.5%), followed by homework (48.3%), report writing (43.7%), and exam preparation (41.7%). Although students reported perceived efficiency gains, concerns persisted about reduced analytical engagement and technological dependence (84.1%). Ethical concerns centered on dependence, authenticity, and data privacy, while institutional responses pointed to the need for formal training (96.7%) and clearer guidance. Based on this triangulation, we propose a context-bounded interpretive framework suggesting that GenAI’s educational value depends on instructional and governance conditions that preserve autonomy, critical thinking, integrity, and equity.
Mayra Álvarez-Jiménez, Geovanny Cudco, Diego Gamboa et al.· Computers· 0 citations
It is suggested that AI can enhance drafting, revision, and feedback processes, improving coherence, metacognition, and writing confidence, however, these benefits are accompanied by persistent concerns regarding ethical ambiguity, inconsistent policy guidance, and insufficient faculty training.
Samira Dichari, Fadi Jaber· Journal of Education and Tra...· 0 citations
The findings indicate that GenAI adoption in African HEIs is expanding but uneven, concentrated in digitally advanced nations, enhancing personalization, multilingual learning, and research productivity, yet raises ethical concerns about academic integrity.
O. Apata, Peter Oyewole, S. Ajose et al.· Journal of University Teachi...· 0 citations
The Responsible AI Literacy in Education (RAIL-Ed) framework is introduced, developed through a systematic review and qualitative framework analysis of 67 studies (2023-2025), grounded in critical, pragmatist, sociocultural, and human-centered traditions.
S. Hossain, S. Ahmadi, Leqi Li et al.· 0 citations
This scoping review examines how generative artificial intelligence (GenAI) tools such as ChatGPT are used in educational contexts and how their use influences students’ critical thinking. A systematic search of four databases (Scopus, Web of Science, ERIC, and IEEE Xplore), followed by screening and eligibility assessment, identified 25 empirical studies published between 2022 and 2025. The findings show that GenAI supports critical thinking when embedded in structured learning activities that require students to question, evaluate, and revise AI-generated content. Tasks such as debate, error detection, and guided writing revision consistently promote analytical reasoning and reflection. In contrast, unstructured use often leads to passive acceptance of AI outputs and reduced cognitive engagement. To account for this divergence, the review proposes the Task–Mediation–Internalisation (TMI) model, which conceptualises AI as a mediational stimulus within a Vygotskian developmental process. Within this system, critical thinking emerges through socially mediated interaction and potentially becomes internalised through repeated meaningful engagement. The findings highlight the importance of designing AI-supported learning environments that position AI outputs as objects of reasoning rather than sources of answers.