Healthcare, smartphones, and the carbon footprint

Smartphones are widely used by physicians and patients. The carbon footprint of healthcare devices is poorly documented. Physicians report an average daily smartphone usage of 1 – 5 h for activities, such as reviewing diagnostic information, capturing patient photographs, conducting telehealth consultations, and advancing their medical education. Meanwhile, patients generate billions of daily queries on Google and millions on ChatGPT, trends likely to increase as artificial intelligence (AI)-driven search engines and large language models (LLMs) become more sophisticated and accessible. To explore the associated environmental impact, we evaluated the average lifetime carbon emissions linked to smartphone usage and the energy costs of manufacturing selected smartphone models. Our data were sourced from publicly accessible databases, corporate 10-K statements, and corporate social responsibility reports available on company websites. We then validated these findings by using four types of LLMs, including ChatGPT, Gemini, Claude.ai, and Meta AI. We found that all LLMs produced carbon emission estimates that differed from those reported in the companies’ official corporate literature. In an era of rapid AI adoption, establishing reliable environmental metrics is essential for informed decision-making and responsible technology use.