By James Arogundade

In Nigeria and globally, the conversation around artificial intelligence (AI) often focuses on its impact on jobs or its use in everyday tech. But a recent study sheds light on a far more critical application: using AI to fight biodiversity loss and manage global health risks.

A new paper, “Augmenting Geospatial Data with Large Language Models Using Compositional Attention for Improved Avian Mobility Tasks Prediction” by Dr. Kehinde Owoeye, a researcher at the National Engineering Laboratory in the United Kingdom, explores how Large Language Models (LLMs)—the technology behind chatbots like ChatGPT can be used to forecast the movement of migratory birds. This study, published by Springer Nature, has real-world implications for Nigeria and the world.

Reducing Biodiversity Loss

Nigeria, with its rich and diverse ecosystems, is a key location for migratory bird species. These birds play a vital role in our environment, from pollinating plants to controlling pests. However, their migration patterns are being disrupted by climate change and human development via the proliferation of critical infrastructures such as wind turbines, high rise buildings, high tension wires, putting both the birds and our environment at risk.

Dr. Owoeye’s research shows that by combining traditional geospatial data (like movement coordinates, weather and elevation) with information from LLMs, we can create more accurate predictions of bird movements. For example, an LLM can provide expressive details about a location, such as the presence of farmlands, urban development, or even tourism, which are not captured by standard geospatial data when used alone.

This combined data gives us a powerful tool to predict and protect nature’s ecosystem. By forecasting the migration states of endangered species, we can make smarter decisions about how to manage critical infrastructure like wind turbines to reduce bird deaths.

Protecting People, Animals, and the Environment

The research also has significant implications for global health. The “One Health” initiative recognizes that the health of humans, animals, and the environment are interconnected. We’ve seen this firsthand with infectious diseases like avian flu, which often originate in animals and spread to humans.

Migratory birds can be vectors for these diseases. Dr. Owoeye’s paper demonstrates that by forecasting how long a bird stays in a particular place, we can map potential disease hotspots. This allows for proactive measures to minimize contact between wild birds and humans, helping to prevent future outbreaks before they happen.

A Sustainable Approach to AI

A major concern with using large AI models is their significant carbon footprint. The research tackles this head-on by proposing an efficient method to reduce the number of queries to the LLM. By only querying for data at key points in a bird’s journey, the system minimizes energy consumption, making the entire process more sustainable.

This is a crucial point for a developing country like Nigeria, where electricity infrastructure and environmental sustainability are pressing issues. The paper shows that we can leverage powerful AI tools without contributing to the very climate problems they are helping to solve.

Bridging the Gap: From Lab to the Field

This study demonstrates AI’s potential beyond commercial applications as a crucial ally in preserving biodiversity and safeguarding our collective future. The next phase according to the paper, involves collaborating with conservation practitioners to implement this solution in real-world environments where it can deliver meaningful impact.