From predictive analytics to AI-powered diagnostic tools, AI is rapidly revolutionizing healthcare. Emerging technologies promise breakthroughs in disease prevention, early detection, and access to care. However, while these systems were built to benefit large groups of people, the unfortunate reality is that when equity isn’t hardwired into these algorithms, AI will only continue to deepen the very disparities it purports to solve. 

For minority and medically underserved communities, the stakes are even higher. We have already seen how biased algorithms can misread data, underestimate the severity of an illness, and perpetuate historical inequities. For example, a 2024 analysis found that several widely used hospital models systematically underassessed the severity of illness for Black emergency department patients, potentially delaying access to lifesaving interventions. 

Still, these inequities extend far beyond algorithms, rooted in the very infrastructure powering AI. Recently, in South Memphis, Elon Musk’s xAI constructed a supercomputer data center in the historically Black neighborhood of Boxtown, where unpermitted methane-fueled turbines have been linked to spikes in air pollution and worsening respiratory issues amongst residents. 

At the same time, we are witnessing AI’s potential to solve some of public health’s most wicked problems. AI-assisted retinal imaging, for example, has accelerated early detection of diabetic retinopathy — a condition that disproportionately affects Black and Hispanic populations. Similarly, AI-driven geospatial modeling during the COVID-19 pandemic helped identify testing deserts in underserved areas—particularly rural communities— enabling local health departments to deploy mobile clinics and save even more lives. These stories are a powerful reminder that when AI is designed with equity at the center, it can not only minimize harm, but actively create pathways to better outcomes for historically underserved communities. 

The challenges and opportunities of artificial intelligence are clear. We are at the nexus of technology and public health, where the choices we make will determine whether these systems will drive equity forward or further entrench existing disparities. To fulfill the promise of AI’s potential, we need to ensure that the voices, needs, and realities of marginalized communities guide how these technologies are being built and deployed. Because at the end of the day, innovation without equity isn’t progress — it’s exclusion. 

This is just the beginning of a much larger conversation. 

Subscribe for part two of a new FYH News series exploring the intersection of AI, health, and minoritized communities. Over the next few weeks, I’ll be sharing additional context, insights, and resources that will explore how this technology can be shaped to empower communities, amplify underrepresented voices, and reimagine what equitable healthcare looks like. 

References:

Chang, T., Nuppnau, M., He, Y., Kocher, K. E., Valley, T. S., Sjoding, M. W., & Wiens, J. (2024). Racial differences in laboratory testing as a potential mechanism for bias in AI: A matched cohort analysis in emergency department visits. PLOS Global Public Health, 4(10), e0003555–e0003555. https://doi.org/10.1371/journal.pgph.0003555

Das, V., Zhang, F., Bower, A. J., Li, J., Liu, T., Aguilera, N., … Tam, J. (2024). Revealing speckle obscured living human retinal cells with artificial intelligence assisted adaptive optics optical coherence tomography. Communications Medicine, 4(1), 1–10. https://doi.org/10.1038/s43856-024-00483-1

Smith, E. R., & Oakley, E. M. (2023). Geospatial Disparities in Federal COVID-19 Test-to-Treat Program. American Journal of Preventive Medicine, 64(5), 761–764. https://doi.org/10.1016/j.amepre.2023.01.022