A Journal of Investigative Dermatology study finds AI diagnostic tools demand significant energy and water — burdens that fall hardest on under-resourced communities.
A single AI model, trained once, can consume as much electricity as a transatlantic flight. That figure sits quietly behind every dermatology clinic that has adopted automated image analysis or AI-assisted documentation.
The Journal of Investigative Dermatology recently examined AI's expanding role in skin care — and the costs that rarely appear in the clinical conversation.
What the research found
AI tools are now used across three areas of dermatology, the journal reported: diagnostic image analysis, clinical documentation, and patient communication. Each application draws on large language or vision models that require substantial computing infrastructure to build and maintain.
Training those models demands significant electricity. The journal noted that data centre energy use has risen in direct proportion to AI adoption across medicine. Cooling that infrastructure adds a second burden: water consumption, an often-overlooked variable in sustainability discussions.
The research identified a distributional problem at the centre of this. Environmental costs, the journal found, fall disproportionately on resource-constrained communities — the same communities that are often furthest from the clinical benefits AI promises to deliver.
Why this matters for the albinism community
People with albinism carry a measurably higher risk of skin cancer, and access to consistent dermatological monitoring is already uneven across geographies. AI-assisted diagnostics hold genuine promise for extending that monitoring — earlier detection, lower cost per screening, broader reach.
But the Journal of Investigative Dermatology study places a question beside that promise. If the infrastructure powering these tools concentrates environmental strain in low-resource regions, the communities with the most to gain may also absorb the most to lose.
The journal described this as an urgent priority for the field — not a distant concern, but one that belongs in the same conversations as clinical accuracy and deployment equity.
The study did not propose a single resolution. It named the gap: dermatology has been largely silent on AI's environmental footprint, even as adoption accelerates.
Keywords
Core topics and entities mentioned in this summary.