AI in Combating Poverty and Economic Inequality

Background and Purpose: Global poverty and economic inequality remain persistent and complex challenges, further intensified by the COVID-19 pandemic, inflation, and structural disparities. In this context, artificial intelligence (AI) has emerged as a powerful tool offering the potential to identify at-risk populations, optimize public resources, and expand access to critical services such as healthcare, education, and credit. However, these technological advances also raise significant concerns around fairness, surveillance, and algorithmic bias. This review explores the dual role of AI in both alleviating and potentially exacerbating economic inequalities. It critically examines how AI can be leveraged to promote economic justice while highlighting ethical risks that may reproduce or even deepen existing disparities, especially among marginalized groups defined by race, gender, and disability.

Methods: Using an interdisciplinary approach, this review synthesizes findings from empirical research, global case studies, and theoretical perspectives across economics, social work, and data ethics. An intersectional framework guides the analysis of AI applications in four key domains relevant to poverty alleviation: predictive analytics, resource optimization, workforce development, and financial inclusion. In addition, the review examines recent policy developments—such as the European Union’s Artificial Intelligence Act—and draws on perspectives from technology, law, and social justice to evaluate emerging best practices for implementing AI in ways that are both ethical and equitable.

Results: AI is transforming the delivery of social services through enhanced poverty mapping, improved targeting of cash transfers, alternative credit scoring, and adaptive learning and employment platforms. Notable advances include the use of satellite data for rural poverty detection, AI-powered job matching for low-income workers, and fintech innovations that extend credit access to the unbanked. However, the analysis also reveals consistent patterns of harm. AI systems trained on biased data have led to discriminatory hiring, credit scoring, and social service eligibility decisions. These risks are magnified in low-resource or high-stakes environments where transparency and oversight are limited.

Conclusion and Implication: While AI can be a catalyst for poverty reduction, its success depends on inclusive design, robust oversight, and interdisciplinary collaboration. The review emphasizes the need for fairness audits, transparent algorithms, community engagement, and co-creation of solutions with affected populations. Social work professionals, policymakers, and AI developers must jointly ensure that AI interventions do not reinforce systemic inequities but instead contribute to a more just and resilient social welfare infrastructure. As AI technologies—including large language models—continue to evolve, embedding ethical safeguards and equity considerations from the outset is essential to realizing their full potential in advancing economic justice.