This systematic review explores how artificial intelligence (AI) supports context-based vocabulary learning in English as a Foreign Language (EFL) and English as a Second Language (ESL) contexts. Guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 framework, the review synthesizes evidence from 65 peer-reviewed empirical studies published between 2010 and 2025. The findings show that AI can enhance vocabulary learning and learner motivation, particularly when vocabulary is learned through meaningful contexts, interactive activities and adaptive feedback. However, existing evidence focuses mainly on short-term learning gains, with limited understanding of long-term retention, transfer to broader language skills and implementation across diverse educational settings. This review also identifies key gaps in the literature, including the dominance of short-term experimental research, the concentration of studies in higher education and limited attention to collaborative and socially mediated learning. Drawing on these findings, this review proposes an integrated framework linking AI-driven personalization, pedagogical design and learning outcomes. Overall, the evidence suggests that AI has considerable potential to support vocabulary instruction, but its effectiveness depends on how well personalization is aligned with sound pedagogy and meaningful learning experiences.