INTRODUCTION:
Digital health and artificial intelligence (AI) interventions are increasingly promoted as transformative tools for addressing persistent health inequities across diverse populations and health systems. Applications such as telemedicine, mobile health (mHealth), AI-driven risk stratification, decision-support systems, and natural language processing have shown potential to improve access, personalize care, and enhance health service delivery in primary care, chronic disease management, mental health, and public health surveillance. However, growing evidence also highlights the risk that these technologies may perpetuate or exacerbate inequities due to algorithmic bias, the digital divide, and the underrepresentation of marginalized populations in data and design processes.

METHODS:
This comprehensive review synthesizes evidence from 50 highly relevant studies identified through a structured search from sources including PubMed and Semantic Scholar. Twenty targeted searches across eight thematic clusters were conducted, covering equity frameworks, population-specific applications, barriers, ethical considerations, and methodological approaches. Studies addressing digital health or AI interventions with explicit relevance to health equity were included and narratively synthesized.

RESULTS:
The evidence demonstrates that digital health and AI interventions can improve access to care, patient engagement, and selected health outcomes, particularly in underserved and resource-constrained settings. Telemedicine and mHealth applications consistently reduced geographic and linguistic barriers, while AI tools enhanced diagnostic accuracy and personalized disease management. Nonetheless, substantial barriers persist. Socioeconomic and infrastructural disparities, limited digital literacy, and biased or unrepresentative datasets pose significant risks to equitable implementation, especially in low- and middle-income countries. Many studies reported inadequate measurement and reporting of equity-specific outcomes, limiting robust assessment of long-term impacts on health disparities.

CONCLUSIONS:
Digital health and AI interventions hold considerable promise for advancing health equity, but their benefits are not automatic. Intentional equity-centered design, community co-creation, robust bias mitigation strategies, and standardized equity-focused evaluation frameworks are essential.