Washington: Environmental organisations around the world are increasingly using artificial intelligence to protect wildlife, monitor forests and improve climate research. However, they are also raising concerns about the growing environmental costs of the data centres that power the technology.
A Reuters report published on Saturday found that all 15 environmental, conservation and development organisations interviewed were using AI in some form. At the same time, they expressed concerns about the electricity, water and other resources required to operate large data centres.
The growing use of AI has created a difficult balance for environmental organisations. While the technology can help them process information more quickly and improve conservation work, its expanding infrastructure can also increase pressure on natural resources and contribute to greenhouse gas emissions.
In Cambodia, Conservation International is using AI to identify possible wildfire hotspots through satellite monitoring. The system sends alerts to community forest patrols, allowing them to respond more quickly to potential fires in remote areas.
Ali Swanson, the organisation's AI lead, acknowledged the environmental costs of the technology but highlighted its potential to support conservation efforts. The system demonstrates how faster access to information can help communities protect forests and respond to environmental threats.
In Kenya, Save the Elephants is using machine learning to improve wildlife surveys and study elephant communication. The organisation recently used the technology to analyse approximately 200,000 images collected during an aerial survey of the country's largest protected area.
According to its chief executive, Frank Pope, nine people previously needed around nine months to examine the images manually. AI can now complete similar analyses within hours.
The organisation also reported that earlier manual surveys had failed to identify approximately one third of the elephants, half of the zebras and giraffes, and three quarters of the oryx recorded through its improved analysis. The findings highlight the potential of better image analysis to provide more accurate information about wildlife populations.
The Jane Goodall Institute is also using AI to preserve important conservation research. In Tanzania's Gombe National Park, the organisation is digitising decades of handwritten records documenting chimpanzee behaviour.
The historical material includes local Kiha language content and scientific shorthand, making the digitisation process difficult. A project that was initially expected to take between 13 and 15 years is now estimated to require approximately two years with the help of an AI tool developed in collaboration with Amazon Web Services.
However, conservation groups are increasingly concerned about the environmental footprint of the infrastructure supporting the technology.
Data centres contain large numbers of computer servers that process and store information. They require substantial amounts of electricity and cooling, while some facilities also place significant demands on local water supplies.
The International Energy Agency estimates that global data centre electricity consumption could reach approximately 945 terawatt hours annually by 2030. This would represent slightly more than Japan's current annual electricity consumption. Artificial intelligence is expected to be the main driver of this increase, alongside other digital services.
Water consumption is another growing concern. Estimates cited in the report suggest that the annual water footprint associated with data centres could increase from approximately 560 billion litres in 2023 to 1.2 trillion litres by 2030. These estimates include water used beyond the facilities themselves, including in electricity generation and semiconductor manufacturing.
The expansion of data centres also raises concerns about land use, electricity infrastructure and greenhouse gas emissions. The environmental impact partly depends on how electricity is generated, with facilities relying on fossil fuels contributing to climate change.
Major technology companies, including Microsoft, Amazon and Google's parent company Alphabet, have reported rising emissions amid the expansion of data centres. Amazon has pointed to efforts to improve energy efficiency and increase the use of lower carbon sources of energy. Microsoft and Google did not respond to requests for comment on the concerns raised in the report.
The debate is receiving greater international attention ahead of the United Nations climate conference COP31, scheduled to take place in Antalya, Türkiye, from November 9 to 20.
Türkiye announced the Antalya Artificial Intelligence Pledge on September 21, 2026, in collaboration with the International Telecommunication Union. The voluntary initiative encourages responsible AI development while addressing the environmental impact of the infrastructure supporting the technology.
Türkiye is seeking wider support from governments, businesses and other organisations during the COP31 process.
The World Resources Institute has welcomed the initiative but called for stronger accountability measures. It has urged technology companies to disclose their energy and water consumption, use clean energy and consider the effects of data centre construction on local communities.
United Nations Secretary General António Guterres launched the AI Environmental Transparency Initiative in June 2026, calling on major AI companies to disclose the environmental impact of their systems, including carbon emissions, water consumption and land use.
He has also urged companies to commit to powering all data centres with renewable energy by 2030. The initiative aims to improve transparency and ensure that communities have access to information about the environmental consequences of expanding digital infrastructure.
Meanwhile, technology companies and environmental organisations in the United States are discussing ways to establish safeguards for data centre development.
A coalition known as the Responsible Data Center Collaborative includes major technology companies such as Google, Microsoft, Amazon, Meta, Oracle, Nvidia and Anthropic, alongside environmental organisations including the World Wildlife Fund, Sierra Club and Natural Resources Defense Council.
The group began discussions in September 2026 and is expected to continue working towards a proposed framework through February 2027. The discussions are expected to cover energy use, water resources, infrastructure development, community involvement, transparency and accountability. However, a final agreement has not yet been established.
Environmental organisations are also exploring ways to reduce the resources consumed by their own AI activities.
The World Resources Institute is limiting applications that require significant computing resources, avoiding unnecessary repeated processing and storing previously generated results. The Nature Conservancy is using cameras equipped with AI technology aboard commercial fishing vessels to analyse catches without transferring large amounts of video to distant data centres.
Greenpeace International, meanwhile, has adopted guidelines emphasising transparency and human oversight while questioning the need for continued expansion of large data centres.
As AI becomes increasingly important in environmental research and conservation, organisations face the challenge of ensuring that the technology does not create additional environmental problems while helping solve existing ones.
The growing debate highlights the need for responsible innovation, greater transparency and stronger cooperation between governments, technology companies and environmental groups to ensure that the benefits of AI do not come at an unacceptable cost to the planet.