Artificial Intelligence in Global Health: Opportunity, Capacity, and the Path Forward
March/April 2026 | Volume 25 Number 2
Artificial intelligence (AI) is rapidly transforming health research and practice, offering new tools to analyze data, improve diagnostics, and strengthen health systems. In my own work and personal life, I use AI tools regularly across a wide range of tasks, and I believe our staff, grantees, and trainees should become familiar with them and use them when appropriate. These tools are already changing how research is conducted, analyzed, and communicated, and those who learn to use them effectively will be better positioned to advance science and improve health. In my talks with early-career colleagues, I warn them that AI won’t take their jobs, but someone who knows how to use AI might.
Read recent commentary on global health research issues from current and immediate past directors of the Fogarty International Center.
The potential impact of AI may be especially significant in low- and middle-income countries (LMICs), where shortages of trained health professionals and limited infrastructure constrain access to care. In such settings, AI has the potential to extend the reach of health systems in new ways. For example, algorithm-driven care models can support frontline health workers in diagnosing and managing common conditions. AI-assisted interpretation of imaging studies, such as chest X-rays or ultrasound, can help address shortages of radiologists. AI tools may also support patient counseling and education, providing tailored information and mental health support in settings where providers have limited time. While these approaches are still evolving, they illustrate how AI could help bridge gaps in human resources and expand access to care.
At the same time, recent work by my colleagues and me highlights both the promise and the challenges of AI in global health. In an analysis of the NIH portfolio, we found that just over 5% of NIH AI-related projects focus on LMICs. This is striking, given that many of the most compelling use cases for AI are directly aligned with global health priorities.
This imbalance matters from a perspective of fairness yet also practicality and precision. AI models trained primarily on data from high-income settings may not perform well when applied elsewhere. Ensuring both accuracy and relevance requires meaningful inclusion of data from LMICs, along with the capacity to analyze and apply those data locally. This is fundamentally a capacity issue. It is not enough to deploy AI tools in LMICs; we must invest in the people, institutions, and data systems needed to develop and adapt these tools in their own settings. Without such investment, AI risks being developed in one context and applied in another, where it may not perform as intended.
In discussions with colleagues and trainees from around the world, there is strong interest in using AI to address pressing health challenges. At the same time, access to data, training, and computational resources remains uneven. Expanding opportunities for researchers in LMICs to participate in AI development and evaluation will be critical, not only to ensure that tools are appropriate for local contexts, but also to foster innovation that can benefit health systems globally.
The growing use of AI also raises practical considerations related to cost, infrastructure, and sustainability. Some AI models require substantial computational resources, with implications for energy use and feasibility in resource-constrained settings. These concerns are particularly relevant in LMICs, where electricity and computing capacity may be limited. This has led to increasing interest in more efficient, “frugal” approaches to AI, developing models that are not only effective, but also affordable and energy-efficient. As in many areas of global health, innovations designed for resource-constrained settings may ultimately prove to be more scalable and sustainable for all.
NIH’s
Harnessing Data Science for Health Discovery and Innovation in Africa (DS-I Africa) program provides one example of an approach that seeks to address these challenges. By investing in data science capacity, supporting African investigators, and fostering collaborative networks, the program is helping to ensure that AI-enabled research is grounded in local expertise and priorities. This includes not only generating and curating data but also training researchers who can develop and apply analytic tools in their own contexts.
Such efforts reflect a broader principle that has long guided Fogarty’s approach to global health research: investing in people and partnerships is essential for achieving lasting impact. In the era of AI, this principle is more important than ever. Building capacity in data science, including representative data, and supporting local leadership will be critical to realizing the potential of AI to improve health outcomes globally.
Artificial intelligence has the potential to accelerate progress in global health, particularly in settings where human resources are limited. But realizing that potential will depend on whether we invest in the capacity, data, and partnerships needed to ensure that these tools are effective and accessible. Done well, AI can help extend the reach of health systems and improve health for all. Or, as I have told our staff, AI will take your job . . . not away, but to the next level.
The ideas expressed here are my own, and I take full responsibility for the content. Consistent with the message of this column, I used artificial intelligence tools to assist in its drafting and editing.
More information
Updated April 22, 2026
To view Adobe PDF files,
download current, free accessible plug-ins from Adobe's website.