Artificial Intelligence Supported Communicative Language Teaching and Its Impact on Speaking Proficiency among Disadvantaged Learners

Students with low socio-economic background are usually left low-proficient in communicating in English even after many years of formal teaching, this is because of the accuracy-based and examination-focused ELT practice which limits meaningful interaction and use of communicative language. The given work will investigate the shortcomings of the classical form-oriented pedagogy, as well as assess the performance of an AI-based communicative ELT strategy in developing speaking competence in low-skilled learners. The current ELT system mainly focuses on grammatical correctness, memorizing and written tests, provides little individualized feedback and insignificant real oral practice. The proposed study aims to overcome these issues by suggesting an instructional model that is based on AI support and learner-centered teaching that incorporates adaptive speaking prompts, automated formative feedback, and AI-mediated conversational simulations complemented with task-based and collaborative exercises. The outcomes of the pre-test/ post-test design, which included low-proficiency learners, indicate that speaking performance has improved significantly, with an increase in the means of speaking scores by 23.17, the fluency measures have improved by 28 and the level of participation by the learners has also increased by 26 to 71 percent. Moreover, self-reported communication anxiety also reduced by 31% which was an improvement of confidence and readiness to communicate because of low-stress AI practice settings adjusted to a particular person. The study has limitations, however, the moderate sample size and contextual focus which can influence the generalizability and lacks the evaluation of long-term retention of communicative gains after the intervention period.