Environmental groups embrace AI, even as they warn of data center impacts

Environmental groups are increasingly leveraging artificial intelligence for climate data analysis, forest fire tracking, and wildlife count, despite acknowledging the technology's environmental impact. AI can aid in combating climate change and protecting natural resources, according to Ali Swanson, AI lead at Conservation International. However, environmental groups express concerns about the environmental consequences of data centers powering AI technology, which consume vast amounts of electricity and water.

Fifteen major environmental organizations responded to Reuters' inquiries about their AI usage, all acknowledging its application and voicing apprehension over data centers' environmental impact. The dilemma of balancing AI's benefits with its risks is expected to draw attention at the upcoming UN COP31 climate conference, where Turkey will propose a pledge for sustainable AI development and deployment.

UN Secretary-General Antonio Guterres is urging major AI companies to disclose the environmental impact of their systems. Researchers estimate that global data center footprints could double by 2030, consuming approximately 945 terawatt-hours of electricity annually and producing carbon dioxide equivalent to Japan's annual electricity use. Water consumption in data centers is projected to rise from 560 billion liters in 2023 to 1.2 trillion liters by 2030.

Technology giants, including Microsoft, Amazon, and Google, have reported emissions linked to data center expansion, raising questions about their commitment to their climate goals. Amazon claims its AI products are among the most sustainable due to low-carbon energy purchases and efficient power and water usage. Microsoft and Google did not respond to requests for comment.

Many organizations believe AI's advantages currently outweigh its costs, as the technology facilitates work that was once challenging, costly, or impossible. For example, Kenya's Save the Elephants uses machine learning for population surveys and elephant communication studies, significantly reducing data review time from nine months and nine people to hours.

Similarly, the Jane Goodall Institute, founded by conservation icon Jane Goodall, employs AI to digitize hand-written field notes, transforming the project from a 13-15 year timeline to approximately two years. Goodall herself supported AI's use in conservation science, provided humans retain control over its application. Some organizations are also exploring ways to mitigate AI's environmental impact, such as limiting resource-intensive applications and using AI-powered digital cameras on fishing vessels to reduce energy use and data-transfer requirements.