AI Data Centers, Water Use, and Maternal Health

Artificial intelligence (AI) data centers are an underrecognized health concern because their rapid growth increases demand for water-intensive cooling and electricity. This narrative review examines AI data-center water use as an upstream determinant of health for pregnant, postpartum, and lactating people. It synthesizes evidence across infrastructure, environmental health, and maternal-child-health literatures to identify plausible pathways and policy implications. A structured search found that data centers can require water, while pregnancy and lactation heighten vulnerability to dehydration and heat, and household water insecurity is linked to poorer health outcomes. Although no direct epidemiologic studies connect AI data-center water use to maternal or infant outcomes, evidence supports pathways through water insecurity, dehydration, thermal strain, hygiene constraints, stress, and cumulative burden, especially in drought-prone settings. AI infrastructure should be included in environmental and maternal-child health surveillance using a precautionary public health approach.