Source: Environment+Energy Leader

AI’s rapid growth is driving significant environmental and social costs. Studies highlight that training large AI models, such as Meta’s Llama-3.1, generates pollution equivalent to 10,000 cross-country car trips, while data centers consume vast quantities of water and energy. These operations are linked to air pollution, causing health risks and exacerbating inequalities in low-income communities near power plants. By 2030, the public health burden of AI-related emissions could rival the total emissions of all vehicles in California. Additionally, the social costs of AI include poor working conditions for data labelers in low-income countries and the perpetuation of bias in algorithms. Despite these challenges, regulatory bodies and tech companies are beginning to address sustainability concerns, with efforts like Europe’s Climate Neutral Data Center Pact and calls for comprehensive reporting standards. As AI continues to expand, balancing innovation with environmental and social responsibility is paramount.