This research scrutinizes the hurdles and prospects associated with the use of AI-driven personalized learning in Indian English Language Teaching (ELT) classrooms. The paper maps out the interplay of policy factors, changes in technology, and the realities existing in the classroom that support or hinder adoption through a systematic synthesis of secondary data which included published studies, government reports, EdTech evaluations, and large-scale pilot findings, among others. Infrastructural weaknesses (limited access to devices and internet) are among the major challenges pointed out, as well as linguistic plurality and the limitations of standard speech and language models being furthered by accents, the gap in teacher readiness for AI-based instruction, and the ethical issues in relation to student data, bias and assessment validity. On the other hand, a research study on the use of adaptive learning programs along with trials carried out in a controlled environment has highlighted the existence of great opportunities: scaled differentiated remediation, more low-stakes speaking and writing practice, richer formative feedback, and analytics that augment the teacher's role when systems are adapted locally and contained within consistent pedagogy. The study suggests a practical implementation framework consisting of equity diagnostics, curriculum alignment, localized model development, teacher co-design and human-in-the-loop quality assurance. The paper ends by stating that AI personalization can significantly supplement the communicative ELT goals in India only if its deployment is pedagogically based, locally adapted, ethically managed and backed by continuous teacher support.
Clive et al. (Wed,) studied this question.