Tech-giant Microsoft’s innovative “AI for Earth” initiative provides research teams with a powerful toolset, including artificial intelligence and cloud-based computing, to tackle significant environmental threats, ushering in a transformative era for conservation science.
AI environment: With growing concerns over the fate of our planet, artificial intelligence (AI) has emerged as an instrumental ally in the global battle against environmental degradation and climate change. Tech behemoth Microsoft, through its initiative “AI for Earth,” is channeling the profound potential of AI into the hands of those who are striving to counteract the daunting environmental challenges we face today.
Microsoft’s “AI for Earth” grant program furnishes beneficiaries with access to advanced AI and cloud computing resources, in a bid to solve multifaceted environmental issues. With researchers grappling with increasingly complex data sets and problems, Microsoft’s cloud-based data storage, management, and analysis platform offer a pivotal stepping stone towards providing solutions for climate change, biodiversity protection, improving agricultural yields, and reducing water scarcity.
Among the beneficiaries are research teams delving into innovative applications of AI to further their environmental missions. For instance, game theory is being applied to anticipate poaching patterns, and machine learning is employed to measure the effectiveness of conservation interventions based on animals’ acoustic activity. Even the realm of social media is mined to ascertain the distributions of specific species.
Advancements in technology have amplified the quantity and quality of information available to scientists, enabling them to accurately gauge the health and behavior of ecosystems and the threats they face. Cryptic cameras, acoustic sensors, satellite imagery, and citizen science apps are among the arsenal of tools that conservationists now wield. In addition, AI is being utilized to distill, analyze, and interpret large data sets to monitor ecosystems and predict outcomes.
Computing systems capable of hosting substantial data sets now use AI to categorize data from various sensors deployed by scientists. The AI environment enables researchers to apply modeling results to generate reproducible code and design user interfaces that allow people to monitor natural systems with high accuracy.
Training computer algorithms with a subset of available data now empowers machines to perform tasks, like classifying photos by species or identifying areas containing water or intact forest. This not only accelerates the work process but also allows for accuracy and precision based on previous data collection and human feedback.
AI’s application in conservation science is multifaceted and immense. Fei Fang, an assistant professor at Carnegie Mellon University, heads a group that uses game theory and machine learning to streamline patrol routes to catch poachers. The Portland-based nonprofit Wild Me utilizes AI-infused computer vision to identify individual animals of endangered species, reducing fieldwork and image processing time under the Wildbook open-source project. Matthew McKown and his team at Conservation Metrics in Santa Cruz apply AI and remote sensing to automate wildlife surveys.
AI for Earth’s grant recipients convened recently at the Microsoft headquarters, in a gathering moderated by Mongabay founder Rhett Butler. The attendees represented over 25 institutions, and they consisted of researchers, engineers, and conservation professionals. They partook in classroom sessions to gain proficiency in cloud-based machine learning tools, deep learning, project management, and data visualization.
Since its inception in December 2017, the AI for Earth program has supported over 110 individuals and organizations in 27 countries. Grantees comprise students, universities, private nonprofit organizations, government agencies, land trusts, and more.
Microsoft’s initiative not only enhances the tech-giant’s image in the environmental sector but also opens doors for potential customers. The beneficiaries appreciate the gains AI-related technologies bring to conservation science. For instance, Fang and her students anticipate machine learning tools will accurately predict wildlife poaching sites, whereas Wildbook’s Holmberg aims to automate the image processing using AI.
At Conservation Metrics, the AI environment is employed to understand the impacts of conservation actions, like prohibiting domestic animals on islands, on seabirds and their nesting habitats. The team is initiating the use of AI to analyze acoustic activity data to determine if such measures effectively restore wildlife populations.
The AI for Earth program prioritizes projects that present a question solvable by AI, aiming to nurture a wide range of technology solutions. It prefers applicants who have identified the issue they wish to address and have gathered the required data to tackle it.
The AI for Earth initiative offers an illustrative example of how AI can hasten the processing of terrain, use, and pollution data to monitor water quality and quantity effectively. The participants of the recent meeting exhibited familiarity with AI and machine learning, but their experience with the Azure platform varied.
The AI for Earth initiative seeks to empower researchers and conservationists with the tools they need to create a sustainable future. By offering support to innovative water and agriculture projects and a host of other environmental endeavors, AI for Earth illustrates that AI is not an existential threat to humanity, but rather a game-changer in our collective quest to preserve the planet.
