Our Landmark Study on Generative AI in LMICs Has Been Published in Nature

When OpenAI released ChatGPT in late 2022, it sparked a global surge of interest in Generative AI (GenAI). Early discussions largely focused on economic productivity, framing GenAI as the next frontier of efficiency and growth. Yet innovators from the Global South have been asking a different question: can GenAI be harnessed to address the pressing social and economic challenges faced by low- and middle-income countries (LMICs)?

This question is at the heart of our latest publication in Nature, “Mapping the Potentials and Limitations of Using Generative AI Technologies to Address Socio-Economic Challenges in LMICs.” Over 70 authors contributed insights on the opportunities and risks these technologies may pose, emphasizing the need for AI development in LMICs to remain equitable, inclusive, and just. The study was made possible with support from the Gates Foundation and developed in partnership with leading Global South think tanks, including India’s Digital Futures Lab and Brazil’s ITS Rio. It provides a comprehensive evidence base on how GenAI is being applied today and how it can be responsibly scaled in LMIC contexts.

The research highlights the opportunities GenAI presents for LMICs, the risks associated with its deployment, barriers to building inclusive applications, and a forward-looking agenda for responsible use. It also includes detailed case studies showing how LMICs are translating GenAI momentum into practical applications across critical development sectors, including health, agriculture, education, financial inclusion, and women’s empowerment. Drawing on 36 real-world use cases, the study shows how GenAI is supporting frontline health workers, improving access to agricultural information, enhancing educational tools, and strengthening service delivery in resource-constrained settings.

Many of these use cases date from 2023 and have evolved since, but the findings remain highly relevant and, in some ways, even more critical given the rapid growth of GenAI technologies. Most examples come from Africa, South America, and Asia, highlighting the leadership and innovation emerging from these regions. The study shows that GenAI achieves its greatest impact when solutions are shaped and led by local innovators, whose deep understanding of socio-economic realities, institutional challenges, and cultural contexts ensures technologies meet the needs of the communities they serve.

While GenAI offers significant opportunities to accelerate development, digital power imbalances in LMICs create serious risks. These include data colonialism, extractive practices, algorithmic bias, threats to jobs and livelihoods, and weak protections for data, rights, and digital sovereignty. Systemic barriers such as underfunded local innovation ecosystems and limited representation of Southern voices in global AI governance further constrain impact.

Maximizing GenAI’s potential in LMICs requires more than technological innovation. Its impact is greatest when paired with rights-based safeguards, strong support for local innovation, and meaningful inclusion of Southern perspectives in research, policy, and governance. A just AI future depends on centering the experiences and leadership of LMIC-based innovators, ensuring their voices guide both technology and policy. This approach provides the evidence and perspective needed to turn GenAI’s potential into meaningful, equitable, and sustainable impact.

Read the full article in Nature: https://www.nature.com/articles/s43588-026-00960-8