Advancing Equitable Education with Inclusive AI to Mitigate Bias and Enhance Teacher Literacy

Background/purpose. The integration of artificial intelligence (AI) in education has the potential to address inequalities and enhance teaching and learning outcomes. However, challenges such as AI biases, limited teacher literacy, and resource constraints hinder equitable implementation, especially in contexts like South Africa. This study investigates strategies for inclusive AI adoption, focusing on localized solutions, co-design practices, and ethical frameworks tailored to the region's unique needs, including linguistic diversity and cultural inclusivity. Materials/methods. Using a literature review methodology spanning studies from 2000 to 2024, this research examines global and local initiatives to identify effective practices for AI integration in education. The study emphasizes the importance of localized datasets, culturally responsive AI tools, continuous professional development, and collaborative learning communities. Results. The study proposes a phased implementation model that includes fairness-aware algorithms, diverse datasets, and sustainable infrastructure investments. It highlights the need to adapt global frameworks to local contexts and foster stakeholder collaboration. These strategies aim to address barriers and provide policymakers and educators with practical recommendations for equitable AI adoption. Conclusion. The findings highlight the importance of localized and culturally responsive approaches to AI integration in education. By leveraging diverse datasets, co-design practices, and ethical frameworks, South Africa can create inclusive AI systems that address inequalities and improve learning outcomes. The study offers policymakers, educators, and stakeholders a practical roadmap to ensure context-sensitive and equitable AI implementation in education.