Policy, Risk and Innovation: A Mixed-Methods Framework for Using AI to Foster Inclusion in Marginalized Communities in Bangladesh

Abstract

This study examines how the adoption of artificial intelligence (AI) and the implementation of policies shape inclusive educational outcomes for marginalized learners in Bangladesh, using evidence from Sherpur Sadar Upazilla. A convergent mixed-methods design integrated a student survey (N = 213; seven institutions; March–September 2024) with qualitative data from 37 stakeholders (teachers and policymakers) collected through semi-structured interviews and focus group discussions. Quantitative findings show that AI tool adoption was the strongest predictor of a composite educational outcome score (β = 0.38, p < 0.001), followed by institutional support (β = 0.25, p = 0.01). In contrast, the policy implementation gap—defined as the mismatch between policy intent and on-the-ground delivery—was negatively associated with outcomes (β = −0.12, p = 0.04). Digital infrastructure quality was positively associated with the outcome but was not statistically significant in the multivariable model (β = 0.17, p = 0.12). The model demonstrated strong explanatory power (R² = 0.67; F(4, 208) = 42.3; p < 0.001). Disparity analyses revealed persistent urban–rural inequities in reliable internet access (94.6% vs. 69.7%) and device readiness, with tablet access emerging as a key enabler of advanced AI-supported learning. Qualitative results corroborated three binding constraints: limited teacher AI preparedness, affordability barriers, and trust concerns related to privacy and algorithmic bias. Building on these findings, the paper proposes a policy–innovation framework centered on localized AI toolkits, sustained teacher upskilling, device-access interventions, and enforceable fairness and transparency safeguards to advance equitable learning opportunities.