In the delicate balance between environmental conservation and economic growth, innovative technologies are emerging as powerful mediators. Artificial intelligence (AI) systems developed at King Abdullah University of Science and Technology (KAUST) are revolutionizing how we monitor and protect natural ecosystems, with profound implications for environmental sustainability, economic prosperity, and policymaking. As individuals grow and thrive in the communities and environments to which they belong, these AI systems play a vital role in solving global challenges affecting human capability development.
The Economic Value of Conservation
Environmental conservation is not only an ecological responsibility, but an economic one. By ensuring sustainable resource use, reducing pollution-related costs, and creating green industry jobs, conservation efforts directly contribute to economic stability and growth.
Consider coral reefs: while occupying just 1% of the ocean floor, these underwater ecosystems support an astonishing 25% of all marine species. Their economic value reaches tens of billions of dollars annually, benefiting more than one billion people worldwide.
Yet climate change threatens these vital ecosystems and, consequently, the economic stability of communities that depend on them. In the Red Sea, home to the world’s second-longest coral reef system, conservation efforts are working hard to protect and support critical species. However, traditional conservation approaches require substantial financial and human resources. This creates an urgent need for more efficient solutions.
AI Tools Transforming Conservation Efforts
KAUST’s newly developed AI systems FishNet and MammalNet represent groundbreaking advancements in conservation technology. FishNet and MammalNet leverage large language models (LLMs) to analyze and identify species from uploaded photos or videos, providing crucial insights into ecosystem health and biodiversity. These systems are evidential of the possibilities that human innovation and new technologies can provide in solving global challenges.
FishNet’s dataset contains approximately 100,000 images covering 17,000 species, while MammalNet’s nearly 600 hours of video make it ten times larger than any existing dataset on animal behavior. Both systems continuously improve their predictive accuracy as users add more input data, creating a virtuous cycle of increasingly reliable environmental monitoring.
These technologies serve conservation work in several important ways.
1. Ecosystem health assessment: users can evaluate the condition of natural habitats and monitor changes over time, providing vital data on conservation strategy effectiveness.
2. Democratized expertise: by making species identification accessible to non-specialists, these tools enable greater local participation in environmental projects, including initiatives by the Prince Mohammed bin Salman Royal Reserve and Red Sea Global.
3. Evidence-based policy: the wealth of data generated supports informed decision-making for climate-conscious policies that simultaneously protect the environment and foster economic development.
Human Capabilities in Harmony with Nature
Environmental well-being is an inseparable part of human capability development. Much like a coral reef where every polyp, fish, and microorganism plays a vital role in maintaining the delicate balance of the entire system, human societies are intricately woven into the fabric of their natural environments. Human flourishing cannot be separated from the health of the natural world people inhabit.
Technologies such as FishNet and MammalNet offer new cost-effective and efficient ways to fulfill this aim. These technologies empower local populations to participate in surveying wildlife and evaluating ecosystem health, creating opportunities for meaningful engagement with conservation efforts. The result is a model where natural environments and economic development can thrive in harmony, reinforcing the mutually beneficial relationship between environmental stewardship and human prosperity.
The HCI theme of "Honoring the Ways We Belong" sheds light on the important role that environmental well-being and climate-conscious policies play in the development of human capabilities, as humans thrive in the environments and communities in which they live. The FishNet and MammalNet technologies under development by KAUST provide essential insights on the health of ecosystems, which inform evidence-based policies and practices that reinforce harmony between individuals and the living environment.
Mohammed Elhoseiny is an Associate Professor in the Computer Science Program at King Abdullah University of Science and Technology (KAUST).
Faizan Khan, Xiang Li, Andrew Temple, Mohamed Elhoseiny, "FishNet: A Large-scale Dataset and Benchmark for Fish Recognition, Detection, and Functional Traits Prediction", ICCV, 2023.
Jun Chen, Ming Hu, Darren J. Coker, Michael Berumen, Blair Roberts Costelloe, Sara Beery, Anna Rohrbach, Mohamed Elhoseiny, "MammalNet: A Large-scale Video Benchmark for Mammal Recognition and Behavior Understanding", CVPR, 2023.
Disclaimer: This report has been prepared by the King Abdullah University of Science and Technology (KAUST). The views and opinions expressed in this report are those of the authors and do not necessarily reflect the views, policies, or positions of the Human Capability Development Program (HCDP), the host of the Human Capability Initiative (HCI) conference.
.png&w=384&q=75)
