Understanding AI-Supported Personalised Learning in Research Writing: A Systematic Review of Technologies, Pedagogy, and Learning Outcomes

HRMARS - Artificial Intelligence (AI) has emerged as a transformative force in education, particularly in supporting personalised learning and research writing processes. However, existing studies remain fragmented, with limited synthesis of implementation strategies, pedagogical design, and learning outcomes. This study presents an SLR of AI-supported personalised learning in research writing. 27 studies issued between 2023 and 2026 were examined using descriptive statistics, thematic synthesis, and VOSviewer-based bibliometric analysis within the PRISMA 2020 framework. The findings indicate that generative AI and large language models (LLMs) are the dominant technologies, primarily supporting adaptive feedback, iterative writing processes, and self-regulated learning (SRL). While AI-supported learning is generally associated with improvements in writing performance, engagement, and learner autonomy, the strength of evidence varies across study designs, with limited and inconclusive findings regarding critical thinking and deeper learning outcomes. Four key themes emerged: (1) pedagogical integration of AI in writing processes, (2) AI-driven personalisation and SRL, (3) learning effectiveness and outcome variability, and (4) emerging trends related to generative AI, AI literacy, and ethical challenges. The review concludes that AI functions as a pedagogically mediated system rather than a standalone solution, with effectiveness dependent on instructional design, learner engagement, and responsible use. Future research should prioritise longitudinal designs, diverse educational contexts, and the integration of AI literacy within pedagogical frameworks.