Artificial Intelligence-Driven Cybersecurity in Resource-Constrained Environments: Challenges and Future Directions
Abstract
Digital infrastructure is expanding rapidly in low-income countries, rural communities, and underfunded institutions, broadening the attack surface for cyber threats while local defensive capacity lags behind. Although AI offers transformative potential for cybersecurity, its adoption in these contexts is constrained by limited computing resources, scarce labeled datasets, fragile connectivity, skills shortages, and weak governance frameworks. This paper surveys AI-driven cybersecurity solutions, examining their applicability and adaptation potential for resource-constrained settings. It analyzes prevalent threat categories, evaluates lightweight models and federated learning as deployment strategies, and identifies technical, economic, and governance barriers. Based on this analysis, a layered cybersecurity architecture is proposed, combining edge AI inference, low-overhead anomaly detection, and community-level threat intelligence sharing. A research agenda and policy recommendations are also outlined to support equitable and sustainable adoption in underserved regions.