How Data Centres Can Support AI-Driven Wildlife Protection

AI continues to shape how organisations tackle environmental challenges, with a new study from Google showing its expanding role in wildlife conservation. 

The report, created with the World Resources Institute (WRI), outlines how AI can help protect biodiversity through open data, but also underscores the infrastructure challenges that come with scaling this technology, particularly for the data centre industry.

AI depends on vast computational power and that power sits inside data centres around the world. While AI's potential to support sustainability initiatives grows, the demand it places on global data infrastructure raises fresh questions about energy use, data access and equitable technology distribution.

A global biodiversity framework with a data problem

The study points to a clear issue: fewer than 25% of countries have specific national biodiversity goals that align with the Kunming-Montreal Global Biodiversity Framework.

This agreement, adopted during the 2022 UN Biodiversity Conference, includes 23 targets aimed at halting and reversing biodiversity loss by 2030. Goals range from protecting habitats and reducing pollution to restoring ecosystems and financing conservation.

However, one of the key barriers to implementing these targets is a lack of accessible and relevant data. AI systems, when trained on robust datasets, offer a way to identify trends, monitor ecosystems and support policymaking. Yet, without sufficient data collection and storage infrastructure, progress stalls.

That infrastructure includes the data centres which store and process high-resolution camera trap images, satellite monitoring feeds and mobile-collected observations. These centres must support the growth of open-access databases while managing their own environmental footprint.

Kate Brandt, Chief Sustainability Officer at Google, says: “From the air we breathe to the food we eat, a healthy planet matters to every single one of us.

“For over 10 years, Google and the World Resources Institute have used the latest technology to protect our planet. But we need to do more, faster.”

AI applications already shaping conservation outcomes

The Google-WRI report highlights examples where AI and data infrastructure already enable conservation on a global scale. These efforts rely on centralised data storage and global access – areas where data centre capacity and connectivity become essential.

Wildlife Insights, developed by Google in collaboration with conservation groups, offers a public database of 253 million camera trap images across 112 countries, covering 4,292 species. These images, hosted and processed in data centres, help scientists analyse species behaviour and population dynamics using machine learning.

Another initiative, Global Fishing Watch, uses AI to track vessel movements and identify illegal fishing activity. In 2024, authorities in Chile used the platform to shut down unauthorised toothfish operations, resulting in 21 vessels facing penalties. This example shows how real-time data analysis can lead to direct enforcement, but only if cloud computing and storage systems remain fast and accessible.

The report also highlights the citizen science platform iNaturalist, where mobile phone users upload biodiversity sightings. With over 100 million research-grade observations, the platform relies on cloud-based data infrastructure to support uploads, verification and global access.

Kate says: “The report highlights real-world examples of people using this technology as we speak to protect and restore nature around the globe.

“Governments using satellites to monitor the seas and prevent illegal fishing. Researchers using AI to help identify, map and protect endangered species. Indigenous communities equipped with real-time alerts to stop illicit logging on their land.”

Infrastructure, access and equity in AI deployment

For AI to reach its full conservation potential, the report outlines three areas of investment. The first is expanding biodiversity data collection and open-access infrastructure. This requires high-volume storage and high-performance computing, both housed within data centres.

Second is the need for transparent and adaptable AI models that can operate in varied ecological regions. To work beyond North America and Europe, AI models must learn from region-specific data. However, data from underrepresented regions is sparse, and data centres in those areas are often limited in number or capability.

Third is capacity-building: enabling conservation workers to use AI tools effectively. This means not only training and support, but access to high-speed cloud platforms that are stable, secure and easy to use.

As Sara Beery, Assistant Professor of AI and Decision-Making at MIT, says: “People spend a lot of time trying to sell models [but] models are only as good as the data. Data is never a bad investment, and data that can be open-sourced and have mutual and diverse downstream uses; that is the no-regret investment.”

Yet with the rapid expansion of AI models comes a cost. According to the International Energy Agency, data centres account for around 1.5 percent of global electricity use, and this figure could double by 2030. That means data centre operators need to balance increasing AI demand with sustainable operations.

The report also raises concerns about AI knowledge being concentrated in a handful of countries, reinforcing global inequality. This includes access to both AI expertise and the physical infrastructure – such as data centres – that supports AI deployment.

Stephanie O'Donnell, Senior Technology Specialist at the World Bank's Global Wildlife Program, says: “Finding the right people and helping them collaborate, build capacity to problem solve and work together is way more important than the technology applications.”

The study estimates that to meet global goals on biodiversity and climate, as defined by the Sustainable Development Goals and the Kunming-Montreal framework, financing for nature must rise by $500bn annually.

Still, AI can help close that gap – if the supporting systems keep pace. That means investing not only in algorithms but also in the data centres powering them. As Brandt says: “Partnership is key to meeting this opportunity.”