Artificial Intelligence (AI) presents unprecedented opportunities for societal transformation, yet its current deployment often exacerbates existing inequalities. This work introduces the Solidarity-Driven Ethical AI Framework (SDEAF), a novel governance model that embeds principles of social justice, collective responsibility, and ecological sustainability into AI systems. Our framework uniquely integrates distributive justice mechanisms in bias mitigation, participatory transparency protocols, and solidarity-based impact assessments. Through case studies in Global South contexts, we demonstrate how SDEAF enables: 1) Equitable resource allocation through bias-aware agricultural AI, 2) Community-led transparency in public service algorithms, and 3) Sustainable AI deployment monitoring using decentralized IoT networks. The framework’s effectiveness is validated through longitudinal studies showing 40% improvement in algorithmic fairness metrics while maintaining 92% operational efficiency across diverse socioeconomic contexts.
The persistent digital divide excludes students in resource-constrained regions from Computerized Adaptive Testing (CAT), which requires stable internet for real-time, serverbased inference. This paper introduces DemocratiAssess , an edge-native framework that enables high-fidelity, offline adaptive assessment. By combining a novel offline-first architecture with extreme 4-bit integer quantization (INT4) with SIMD-optimized bitunpacking kernels and knowledge distillation, we compress transformer models for local deployment on affordable Raspberry Pi hardware, eliminating cloud dependency. Experimental results demonstrate that DemocratiAssess achieves an average system latency of 38.2 ms (σ = 4.1ms, P95=52ms) and a memory footprint of 150 MB, representing a 97% reduction compared to cloud-based systems, while retaining 96.2% of the baseline model’s psychometric accuracy and reducing measurement disparity by 85.7% through Psychometric-Aware Quantization (PAQ). Three-year total cost of ownership (TCO) is reduced to $90.12 including hardware and electricity. This work establishes the technical and economic feasibility of equitable, personalized assessment in low-connectivity academic environments, offering a sustainable model for educational technology in the Global South.
We introduce Contextualized Edge Pedagogy (CEP), a design framework that reconciles the tension between global AI advancement and localized educational equity in post-conflict environments. Drawing on a 65-question stakeholder survey with 250 visually impaired students—representing 27.9% of the visually impaired student population in Tigray, Ethiopia—we identify evidence-based design requirements and propose a context-aware architecture that lays the ground-work for future intervention studies. Conventional educational technology assumptions—stable infrastructure, individual device ownership, and cloud-dependent architectures—systematically exclude visually impaired students in resource-constrained settings. Our three-tier imputation strategy with self-reinforcing validation achieves 88.3% data completeness (97.8% of applicable cells) while preserving cultural grounding, enabling preliminary design hypotheses for offline-first accessible education. A companion paper reports demographic and comorbidity analyses from the same cohort.