Our mission is to help FGPs close the wage and opportunity gap, and part of that means giving women the tools to speak credibly in rooms where AI infrastructure decisions get made, rooms that have historically excluded people who look like us.

The Han et al. (2024) study’s own recommendation, that “all impacted communities” deserve attention, not just the ones near the data center, echoes Executive Insider’s core belief: invisible disadvantages deserve visible accountability.

What the Article Actually Says

Han et al. (2024) built a rigorous, EPA-model-based analysis showing that US data centers could contribute to about 600,000 asthma symptom cases and 1,300 premature deaths in 2028, with public health costs exceeding $20 billion under a high-growth scenario (Han et al., 2024). The research demonstrates that health costs are unevenly distributed geographically, with the highest county-level per-household health cost about seven times the national average and roughly 200 times the lowest county, meaning some communities absorb a disproportionate share of the harm.

Notably, the study finds that health costs don’t track carbon emissions the way people assume; the same amount of carbon produces the same climate impact everywhere, but the same air pollutants can devastate one neighborhood and barely touch another depending on population density and wind patterns. The authors propose “Health-Informed Computing” (HICO), a framework that factors public health costs into AI training and data center scheduling decisions, plus a recommendation that companies expand their sustainability reporting to include health impacts, not just carbon (Han et al., 2024).

A Responsible AI Leadership Judgment Call

This isn’t opinion or activism; it’s peer-reviewed modeling using the EPA’s own risk-assessment tool. Han and colleagues quantify, county by county, what AI’s infrastructure boom might cost in asthma cases and premature deaths, and the pattern is not random (Han et al., 2024).

For first-generation female professionals (FGPs), this research matters because it reframes AI adoption from a purely technical or productivity question into a leadership judgment call. If you’re advising on vendor selection, sustainability reporting, or technology strategy, you’re now equipped with hard numbers, not vague concerns.

What This Means for You, Right Now

If your organization is deciding whether, where, or how to scale AI infrastructure, this study gives you something concrete to bring into the room. The researchers found the same AI training workload can cost $0.23 million in health impact in one location and $2.5 million in another, purely because of where the electricity comes from (Han et al., 2024). That’s not an abstract environmental talking point; it’s a decision variable.

This matters whether you’re 5 years into your career or 55. Early-career FGPs building credibility in tech-adjacent or operations roles can use this data to ask sharper questions in planning meetings. Mid-to-late-career FGPs moving into executive or oversight positions can use it to push for the kind of “extended energy reporting” the study itself recommends, positioning themselves as the person who sees the whole system, not just the deliverable.

Why This Matters for FGPs (First-Gen Female Professionals)

This research anchors Executive Insider’s AI literacy work in something harder to dismiss than intuition: peer-reviewed, EPA-validated data. Our mission is to help FGPs close the wage and opportunity gap, and part of that means giving women the tools to speak credibly in rooms where AI infrastructure decisions get made, rooms that have historically excluded people who look like us.

The study’s own recommendation, that “all impacted communities” deserve attention, not just the ones near the data center, echoes Executive Insider’s core belief: invisible disadvantages deserve visible accountability. You don’t have to be an engineer to ask, “Whose health is this decision affecting, and did we account for it?”

Your Next Step

The next time your team discusses AI vendors, cloud providers, or infrastructure investment, ask a specific question modeled on this research: what’s the estimated health cost of where this workload gets processed, and has anyone modeled it? That question takes you from passive participant to informed decision-maker, exactly the shift Executive Insider exists to support.

Sources

Han, Y., Wu, Z., Li, P., Wierman, A., & Ren, S. (2024). Health-informed computing: Estimating and addressing the public health impact of data centers. arXiv. https://doi.org/10.48550/ARXIV.2412.06288