Conservation groups use AI to track wildlife and fires, while weighing its footprint

Environmental groups are using AI to analyse wildlife and climate data, even as they warn about the electricity, water and emissions associated with data centres. A Reuters report describes both sides of that trade-off, including tools that speed up conservation work and ways organisations are trying to limit the costs.

Watch Desk analysis

What happened

Reuters interviewed 15 environmental, conservation and development organisations. All said they use AI in some form and expressed concern about the environmental impact of the data centres powering it.

The applications are specific. Conservation International uses AI to assess satellite-detected hotspots and send real-time wildfire alerts to forest patrols in Cambodia. Save the Elephants says AI can analyse aerial survey images in hours. The previous manual review of roughly 200,000 images took nine people nine months; the organisation says human surveys had missed about a third of elephants and larger shares of some other species.

At the Jane Goodall Institute, AI is helping digitise handwritten field notes in Tanzania. A project once expected to take 13 to 15 years is now expected to take about two, the institute says. The Nature Conservancy, meanwhile, processes fishing-vessel video on AI-powered cameras aboard the vessels, rather than sending large volumes of footage to central data centres.

Why it matters

These examples show why conservation organisations may accept AI’s costs: it can make work faster, cheaper or possible at a scale that manual analysis struggles to match. The reported results are organisations’ accounts, not a uniform independent assessment of accuracy or environmental impact.

The costs are not abstract. The International Energy Agency estimates data centres could use 945 terawatt-hours of electricity a year by 2030, while annual water consumption could rise to about 1.2 trillion litres. Those projections put the groups’ practical choices in a larger debate about how AI is powered and where its computing happens.

Our read

This is a more interesting argument than “AI is good” versus “AI is bad”: conservation groups are already using the technology, while trying to make its benefits justify its resource demands. Some are limiting energy-intensive uses or avoiding unnecessary repeat queries; others are moving processing closer to where data is collected.

The work described by Reuters gives readers concrete examples to judge, not a universal verdict. The key question is whether these tools produce conservation gains that would otherwise be missed, and whether organisations can demonstrate those gains without treating energy and water as somebody else’s problem.

What to watch

• Whether conservation groups publish measurable results for AI-assisted surveys and alerts.
• Whether more organisations process data locally to reduce transfers to central data centres.
• What sustainability commitments governments and companies make at COP31, and whether they include transparent reporting on AI’s resource use.

Discussion spark: When conservation groups say AI helps them protect nature, what evidence should be required to show that the benefits outweigh the energy and water costs?

Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.